Global-scale population genetic analysis of Plasmodium falciparum identifies country- and region-specific patterns of malaria parasite adaptation

Global-scale population genetic analysis of Plasmodium falciparum identifies country- and region-specific patterns of malaria parasite adaptation | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Global-scale population genetic analysis of Plasmodium falciparum identifies country- and region-specific patterns of malaria parasite adaptation Nina Billows, Jamille G. Dombrowski, Joseph Thorpe, Leen Vanheer, and 9 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7160640/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract To investigate the global evolution and adaptation of Plasmodium falciparum, we analysed 17,565 isolates collected over three decades from 39 countries. This large-scale genomic study integrates identity-by-descent (IBD) networks, population structure and multi-layered association analyses to explore patterns of selection, transmission and drug resistance. Strong selection signals were observed at known resistance loci (e.g. pfcrt, pfdhps) and in genes linked to immune evasion and drug efflux (e.g. pfABCK1, pfMC-2TM), with signals varying by geography and time. In Southeast Asia, clonal expansion of the pfkelch13-C580Y mutation occurred and multi-layered analysis revealed co-occurrence with mutations in pfarps10, pfrad5, and pfMyoF, supporting a polygenic model of artemisinin resistance. In South America and Horn of Africa, elevated IBD was found in pfKIC7 and pfKIC9, interactors of pfkelch13, suggesting convergent evolution under drug pressure. New data from Brazil and Vietnam enhance resolution of global parasite diversity, highlighting the role of genomic surveillance in malaria control. Biological sciences/Genetics/Population genetics Biological sciences/Microbiology/Parasitology/Parasite genomics Figures Figure 1 Figure 2 Figure 3 Figure 4 BACKGROUND Malaria, a vector-borne disease caused by Plasmodium parasites and transmitted by Anopheles mosquitoes, remains one of the most significant global public health challenges. Among the six Plasmodium species that infect humans, Plasmodium falciparum causes the most severe disease and accounts for the majority of malaria-related deaths, particularly among young children in sub-Saharan Africa. In 2023 alone, global estimates reached 263 million malaria cases and close to 600,000 deaths, underscoring the persistent burden of this disease despite decades of control efforts 1 . While progress has been made through antimalarial treatments and vector control strategies, multiple biological threats have emerged that compromise their effectiveness 2,3 . These include the evolution of drug resistance in the parasite and insecticide resistance in mosquito vectors. In addition, the influence of global warming is increasingly recognised as a major factor shaping malaria transmission patterns, altering seasonality, changing human migration patterns and so potentially expanding the geographic range of the disease 4 . Together, these challenges underscore the need for a coordinated global response to control malaria and progress towards its eradication. A critical component of this response is addressing the biological processes that undermine current interventions. In particular, understanding the genetic diversity and evolutionary dynamics of P. falciparum enables tracking of the emergence and spread of drug resistance. While much of this work has focused on national or regional scales, expansion of efforts to map the evolutionary trajectories of P. falciparum to a global level can anticipate resistance trends and hasten an effective response to waning effectiveness of malaria drug regimens. The evolution of P. falciparum reflects a complex interplay of ecological, biological and anthropogenic factors. In addition, malaria control interventions have imposed strong selective pressures, causing population bottlenecks and regional differences in diversity 5-9 . Regional differences in P. falciparum population structure have been well documented. African populations show high diversity due to intense transmission, while lower diversity is seen in South America and intermediate levels in Southeast Asia. These patterns align with region-specific antimalarial drug use and resistance histories 10-13 . Key resistance markers, such as mutations in pfcrt (chloroquine resistance transporter; chloroquine), pfmdr1 (multi-drug resistance; amodiaquine, chloroquine, mefloquine, piperaquine), pfdhfr (dihydrofolate reductase; pyrimethamine), d fdhps (dihydropteroate synthase; sulfadoxine) and pfkelch13 (Kelch propeller protein; artemisinin), exhibit geographic variation, but current panels remain incomplete 14-18 . These resistance markers include both single-nucleotide polymorphisms (SNPs) and structural or copy number variants, all of which exhibit regional variation in prevalence and pattern 19 . Continued identification of novel mutations is critical for surveillance and guiding treatment 20 . Artemisinin resistance, first emerging in Southeast Asia, now poses a growing threat in Africa, with concerning signs of reduced ACT partner drug efficacy, such as lumefantrine resistance in Uganda 21,22 . Population genomics is a powerful tool to dissect the genetic landscape of P. falciparum and uncover the mechanisms driving its evolution and adaptation. By combining classical statistical techniques and high-resolution genomic analyses, researchers can quantify nucleotide diversity, identify loci associated with drug resistance and population divergence, characterise population structure and genetic relatedness, and unravel complex infection patterns 23-26 . These approaches have significantly advanced our understanding of the connectivity of parasite populations, provided insights into gene flow and admixture, and shown how quickly P. falciparum can adapt in response to selective pressures 27-29 . Importantly, population genetics has provided a framework for inferring the origins and spread of drug-resistant mutations, as well as for analysing diversity in genes encoding vaccine targets, both of which are major contributors to intervention response 30 . The expansion of publicly accessible genomic datasets has further enhanced the scale and resolution of such analyses. Notably, the 2023 release of the Pf7 data resource by MalariaGEN included nearly 20,000 P. falciparum whole genome sequences from 33 countries 26 . A number of research groups have also made substantial contributions to the growing repository of genomic data, including from formerly underrepresented malaria-endemic countries, further enabling large-scale comparative analyses and global surveillance efforts 31-37 . Leveraging the growing wealth of publicly accessible genomic data, alongside newly generated whole-genome sequencing (WGS) data from Brazil and Vietnam, this study employs large-scale SNP-based analyses to approach P. falciparum population dynamics from a global perspective. Specifically, the contribution of genetic variation to regional differences in infection complexity, parasite diversity and the emergence and dissemination of antimalarial drug resistance is investigated. By comparing transmission dynamics and evolutionary pressures across diverse geographic settings, the response of P. falciparum to local intervention strategies can be examined. Importantly, the approach also explores the genomic architecture of resistance, including evidence for polygenic interactions that may drive adaptive responses. These findings reveal how regional drug use and transmission dynamics shape P. falciparum evolution, underscoring the need for region-specific genetic surveillance and treatment policies to effectively manage resistance and support malaria elimination. RESULTS Genomic and Geographic Diversity of P. falciparum isolates After quality control procedures, the final dataset comprised 17,565 P. falciparum isolates from 39 countries across 9 regions (Africa (Central, East, Horn, West, South Central), Asia (South, Southeast), South America, Oceania; Table 1 ). Most isolates originated from West Africa (33.3%) and Southeast Asia (33.0%), with smaller percentages from East Africa (9.1%), South America (8.9%), South Asia (8.8%), Central Africa (2.9%), Oceania (2.0%), South Central Africa (1.5%), and the Horn of Africa (0.5%). Notably, Ghana contributed a significant portion of the isolates (n=2,845, 16.2%), while each of the other countries contributed less than 10% to the total sample set ( Supplementary Table 1 ). High-quality genome-wide variants were identified (3,600,488 SNPs in total) and 1,717,351 (47.7%) biallelic SNPs (population-specific minor allele frequency (MAF)>0.001) met the criteria for further analysis. Most shortlisted SNPs were non-synonymous mutations (57.5%). Overall, SNP diversity was comparable across all regions (median SNP π range: 8.24 × 10⁻⁴ – 1.66 × 10⁻³), with slightly lower estimates for South American samples ( Table 1 ). Population structure and Admixture Multidimensional scaling (MDS) revealed that isolates clustered according to their respective geographical regions and countries ( Fig. 1 ). Significant overlap was observed among the African regions (Central Africa, West Africa, East Africa, Horn of Africa and South Central Africa) as well as among the Asian/Oceanic regions (South Asia, Southeast Asia and Oceania) ( Fig. 1 ). South American P. falciparum isolates display distinct regional structuring, reflecting geographic separation and historical transmission ( Supplementary Figure 1 ). Colombian and Ecuadorian isolates form a distinct cluster that is relatively close to African samples along the primary axis of genetic variation. However, they show increased separation along secondary components that capture additional population structure, highlighting subtle divergence. Brazilian isolates are more dispersed, overlapping with African, Colombian and other South American clusters. In contrast, isolates from Peru, French Guiana, Guyana and Venezuela group closely together to form a separate South American cluster. All South American and African clusters are clearly distinct from those in Southeast Asia, South Asia and Oceania ( Supplementary Figure 1 ). To further investigate the geographical origins of P. falciparum isolates, ancestry inference was conducted using maximum likelihood estimation via ADMIXTURE analysis, based on 974,060 SNPs with a MAF greater than 1%. The optimal number of ancestral populations (K = 10) was determined using cross-validation. The ADMIXTURE results revealed patterns of shared ancestry consistent with known geographic origins. For instance, four dominant ancestral clusters were observed among African isolates, with West Africa containing a distinct ancestral component compared to other African regions ( Fig. 1 ). Similarly, Southeast Asian isolates exhibited four dominant ancestral populations, with shared components also observed with isolates from Oceania, reflected in three overlapping ancestral clusters. Several Southeast Asian samples showed signatures of East African ancestry, supporting hypotheses of both the African origin of P. falciparum and human migration from Africa to Southeast Asia 38 . Isolates from South America and Southeast Asia exhibited more distinct and region-specific ancestry profiles relative to other parts of the world. At the country level, ancestral compositions were generally consistent within national borders, with overlapping patterns observed in neighbouring countries ( Supplementary Figures 2–3 ). In contrast, isolates from Colombia and Ecuador had a unique ancestral composition, aligning with MDS findings and suggesting multiple introductions of P. falciparum into South America during historical periods of colonisation 5 . The Horn of Africa has also exhibited some limited gene flow from South Asia, as well as mainly Central-East African gene flow ( Supplementary Figure 4 ). Multiplicity of Infection The F WS statistic was employed to summarise relative inbreeding and multiplicity of infection (MOI), where higher values (closer to 1) indicate clonality, and lower values (closer to 0) reflect a mixture of unrelated clones. An F WS value greater than 0.95 typically signifies a monoclonal population. Regionally, the highest median F WS values were observed in South America (0.99), Horn of Africa (0.99), Southeast Asia (0.98) and Oceania (0.97) ( Table 1, Supplementary Figure 5 ). These regions also had a higher percentage of samples with F WS > 0.95, indicating low MOI. In contrast, lower mean F WS values and a smaller percentage of samples with F WS > 0.95 were reported for South Central Africa (0.74), Central Africa (0.82), West Africa (0.85), East Africa (0.85), and South Asia (0.90) ( Supplementary Figure 5 ). At the country level, F WS values fluctuated over time which could suggest changes in transmission intensity. Statistically significant changes (P < 0.05) in MOI across year groups were observed in Bangladesh (ANOVA, P = 0.025), Cambodia (P = 0.025), Colombia (P = 0.021), The Gambia (P < 2×10⁻¹⁶), Ghana (P = 6.07×10⁻⁶), Kenya (P = 1.98×10⁻⁸), Laos (P = 2.31×10⁻⁹), Mali (P = 1×10⁻⁴), Myanmar (P = 0.001), Papua New Guinea (P = 0.003), Thailand (P = 9.21×10⁻⁵) and Vietnam (P = 2.33×10⁻⁹) ( Supplementary Table 2 ). In several countries, rising F WS scores over time may indicate declining transmission or increased sampling in areas of lower endemicity. Notable examples include Bangladesh, with a median F WS increasing from 0.89 (range: 0.45–1.00) in 2008–2011 to 0.97 (range: 0.45–1.00) in 2016–2019; The Gambia, from 0.74 (range: 0.25–1.00) pre-2000 to 0.97 (range: 0.43–1.00) in 2016–2019; Ghana, from 0.86 (range: 0.29–1.00) 2008–2011 to 0.89 (range: 0.25–1.00) 2016–2019; and Laos, from 0.97 in 2008–2011 (range: 0.54–1.00) to 0.99 in 2016–2019 (range: 0.49–1.00). In other countries, F WS scores showed variability and modest declines over time, potentially indicating rising transmission intensity. This pattern was observed in Benin, Cambodia, Guyana, Indonesia, Papua New Guinea and Vietnam ( Supplementary Table 2 ). Notably, some countries experienced more pronounced declines in F WS scores in recent years compared to earlier periods. For example, Mali saw a drop from a median F WS of 0.93 (range: 0.34–1.00) in 2016–2019 to 0.73 (range: 0.48–0.99) in 2020–2021. Similarly, Kenya’s scores decreased from a median of 0.98 (range: 0.26–1.00) in 2012–2015 to 0.87 (range: 0.48–1.00) in 2016–2019. These shifts could reflect substantial increases in transmission or a focus on sampling in higher-transmission areas. Genomic Relatedness Identity by descent (IBD) analysis was performed to identify genomic segments shared between P. falciparum isolates that have not undergone recombination, serving as markers of inheritance from a recent common ancestor and providing a measure of genomic relatedness. The fraction of the genome classified with IBD was calculated using 10 kb sliding windows and median IBD fractions (with ranges) were initially summarised by regional groupings ( Supplementary Figure 6 ). The highest IBD fractions were observed in samples from South America (median: 0.150; range: 0.003–0.663), followed by Southeast Asia (median: 0.055; range: 0.001–0.238), the Horn of Africa (median: 0.029; range: 0.001–0.227), Oceania (median: 0.028; range: 0.001–0.212), and South Central Africa (median: 0.012; range: 0.001–0.141). In contrast, the lowest IBD fractions were seen in South Asia (median: 0.002; range: <0.001–0.119), East Africa (median: 0.001; range: <0.001–0.114), Central Africa (median: 0.001; range: <0.001–0.114), and West Africa (median: 0.001; range: <0.001–0.100). These findings are consistent with patterns observed in F WS scores and reflect the impact of higher transmission intensity in these regions. Similar trends were observed for pairwise IBD fractions, with South American samples exhibiting the highest levels of pairwise genomic relatedness ( Supplementary Figure 7 ). Regions with higher IBD fractions may indicate reduced outcrossing within parasite populations, typically associated with low transmission intensity or geographic isolation. Conversely, lower IBD fractions suggest increased outcrossing due to higher transmission and greater genetic mixing between isolates. To further explore this, genomic regions with the highest IBD fractions were examined to identify associated genes and gene products ( Supplementary Table 3 ). A total of 377 high-IBD segments (top 5% of IBD values) were identified, encompassing 173 genes. The greatest number of high-IBD segments was observed in the Horn of Africa and Oceania (n = 125 each), followed closely by East Africa (n = 122), South America (n = 121), South Asia and Southeast Asia (n = 120 each), West Africa (n = 120), Central Africa (n = 119) and South Central Africa (n = 117). Several segments were conserved across multiple regions ( Supplementary Figure 8 ). Gene Ontology (GO) term overrepresentation analysis showed that high-IBD regions across all geographic areas were significantly enriched for genes involved in protein–DNA complex subunit organisation, chromatin and chromosome organisation, and the regulation of transcription, translation, gene expression and biosynthetic processes ( Supplementary Table 4 ). These genes were primarily associated with chromosomal regions (fold change: 8.25; n = 5) and the nucleus (fold change: 1.59; n = 49) and exhibited molecular functions largely related to binding and structural molecule activity. Of particular interest, several high-IBD regions overlapped with loci associated with antimalarial drug resistance. These included pfcrt (in Central Africa, East Africa, Horn of Africa, Oceania, South America, South Asia, and West Africa), pfdhps (in East Africa, Horn of Africa, Oceania, South America, South Asia, South Central Africa, and Southeast Asia), and pfmdr1 (in Central Africa, East Africa, Oceania, and South America), potentially reflecting drug-driven selective pressures. Elevated IBD was also found in regions encoding Kelch13-interacting candidate genes KIC7 (Horn of Africa and South America) and KIC9 (South America), suggesting possible signatures of positive selection. Additionally, genes essential for parasite transmission and interaction with Anopheles mosquitoes, P47 (involved in immune evasion) and P48/45 (linked to gamete fertility), showed high IBD in Oceania ( Supplementary Table 3 ). Population Differentiation To further investigate the genetic differentiation of P. falciparum across global populations, fixation index (F ST ) analysis was conducted to identify highly differentiated genomic sites and assess patterns of divergence between regional subpopulations. A strong positive correlation was observed between genetic differentiation and geographic distance (Mantel r = 0.89, P = 0.001) ( Fig. 2 ). The lowest levels of genetic differentiation were observed within Africa (genome-wide F ST range: 0.009–0.071), as well as between South and Southeast Asia (genome-wide F ST : 0.083). In contrast, the greatest differentiation was observed between Southeast Asia and South America (genome-wide F ST : 0.365), reflecting the large geographic separation between these regions (~18,327 km) and historically low human population movement between them ( Fig. 2 ). Patterns of moderate to high differentiation (F ST > 0.75 and F ST > 0.95) also aligned with geographic distance. The highest number of highly differentiated SNPs (F ST > 0.75) were observed in comparisons involving South America: with Southeast Asia (n = 172, including 41 with F ST > 0.95), South Asia (n = 159, including 10 with F ST > 0.95) and Oceania (n = 82, including 32 with F ST > 0.95), supporting the influence of geographic distance on genotypic divergence. In contrast, no SNPs with extreme differentiation (F ST > 0.95) were identified between South and Southeast Asia, consistent with their relative proximity (average sampling distance: 1,722 km) and long-established human migration. Within Africa, only 14 highly differentiated sites (F ST > 0.75) were identified across five genes: RhopH2 , PF3D7_1361800 , PF3D7_0811600 , Gcalpha , and PF3D7_0526600 ( Supplementary Table 5 ), reflecting the lower overall genetic structure across the continent. A total of 1,346 SNPs exhibited high genetic differentiation (F ST > 0.75) in intercontinental comparisons ( Supplementary Table 5 ), highlighting key genomic regions underlying geographical divergence, including loci implicated in drug resistance and parasite transmission. Noteworthy examples include pfdhfr (pyrimethamine resistance), pfmdr1 (chloroquine and mefloquine resistance), and pfcrt (chloroquine resistance), all of which harbour mutations known to mediate resistance phenotypes. Although no significant GO terms were identified for biological processes or molecular functions, sites with F ST > 0.75 were found within genes localised to cellular components, including the intrinsic component of the external side of the plasma membrane, anchored components of the plasma membrane, the apical complex, cell surface and apical part of the cell ( Supplementary Table 6 ). Genes from multiple families also contained highly differentiated sites, particularly those involved in lipid scavenging, such as the ACS family (n = 38) and ACS10 (n = 34), which have been proposed as drug targets, but are also highly variable 39 . Additional high-F ST sites were observed in genes mediating parasite-host and parasite-vector interactions, including P47 (n = 33), CTRP (n = 20), and Pfs16 (n = 9), or genes implicated in immune evasion, cellular invasion, or gametocyte development, and are considered vaccine candidates 26,40 . Comparable patterns were observed in country-level pairwise F ST analyses ( Supplementary Table 7 ). Interestingly, despite their geographical proximity, Brazil and Ecuador showed a relatively high number of highly differentiated sites (F ST > 0.75), a pattern more likely attributable to small sample sizes rather than true biological divergence. However, it could also reflect both geographic and sociolinguistic barriers to gene flow, including the lack of a shared border, distinct coastlines (Atlantic vs. Pacific) and differing colonial and linguistic histories that may limit parasite movement via human migration ( Supplementary Figure 1 ). Further insights into regional differentiation were obtained using a 'one-against-all' F ST approach. At the regional level, only 30 sites showed high genetic differentiation (F ST > 0.75) and none exceeded an F ST of 0.95 ( Supplementary Table 8 ). Most of these high F ST sites were specific to South American isolates (n = 17), with fewer identified in West Africa (n = 9), Oceania (n = 3) and Southeast Asia (n = 1). These loci were located in genes associated with drug resistance ( pfdhfr , Oceania), regulation of biological processes ( CRK3 , SET1 ), cellular metabolism ( ACS10 , PF3D7_0709700 ), biosynthesis ( PF3D7_0713600 , PAIP1 ), reproduction ( PF3D7_0809600 , GIG ), nutrient uptake ( RhopH2 ), cell localisation ( PF3D7_1440800 , HSP101 ), signalling ( GCbeta ), cell binding ( PF3D7_1410400 ), and interspecies interactions ( P47 ), as well as other genes with diverse functions ( PF3D7_1116800 , PF3D7_1135100 , PF3D7_1442200 ). At the country level, the 'one-against-all' analysis identified 211 high F ST sites, comprising 132 unique SNPs with F ST > 0.75 and two sites with F ST > 0.95 ( Supplementary Table 9 ). Most of these sites were observed in isolates from Brazil, Ecuador, Colombia, Peru, and Indonesia, underscoring localised genetic structure. Several of these high F ST sites overlapped with previously described genes and regions known to be highly variable, such as the SURFIN family, pfcrt , and pfdhfr . Notably, isolates from Brazil, French Guiana, Peru, and Indonesia also carried highly differentiated sites in FIKK4.2 and FIKK10.1 , genes previously implicated in parasite invasion and virulence 41 . Together, these results emphasise region- and country-specific patterns of genetic differentiation in P. falciparum , likely driven by a combination of geographic isolation, local selective pressures, and differing transmission dynamics. Genomic regions under recent positive selection To investigate recent positive selection across the P. falciparum genome, we applied within-population (iHS) and between-population (XP-EHH) analyses based on extended haplotype homozygosity (EHH). For iHS, SNPs exceeding the threshold of (− log10[1 – 2 | ΦiHS – 0.5 |]) > 4.0 were under recent positive selection within single regions ( Supplementary Figure 9; Supplementary Table 10 ). In total, 561 unique SNPs spanning 170 genes met this criterion. The highest number of sites were observed in West Africa (n = 207), followed by East Africa (n = 190), South Asia (n = 169) and Central Africa (n = 144), while fewer than 30 SNPs were detected in other regions. Most of the positively selected sites were found in genes encoding surface antigens or located in highly variable genomic regions, consistent with immune-mediated selection and adaptation for efficient merozoite invasion. These included DBLMSP2 , TRAP , CLAG8 , AMA1 , MSP1 , and members of the SURFIN family. Notably, these genes were enriched in the KEGG pathway pfa05144 (adjusted P = 0.001), which is associated with sporozoite invasion. In addition, 16 SNPs under positive selection were detected in genes involved in lipid metabolism, particularly ACS and ACS7 , across East, West, Central Africa and Oceania. These acetyl-CoA synthetase genes play a key role in scavenging host fatty acids to support parasite growth 39 . Drug-resistance loci also showed signs of selection. For example, a non-synonymous mutation in pfcrt (405600C>T, Ile356Thr) was identified in West African populations, potentially reflecting historical chloroquine pressure or adaptation to current combination therapies. To identify regional selection hotspots, we further prioritised genes containing >5 SNPs exceeding the selection threshold, as well as the top five strongest selection signals per region ( Supplementary Figure 9 ). In African populations, notable hotspots included PF3D7_0809600 (C50 cysteine protease) and PF3D7_1475800 (hypothetical protein) in East Africa; PF3D7_1028000 (uncharacterised, also selected in West Africa) in the Horn of Africa; and PF3D7_0113800 (DBL-containing protein) in South Central Africa. Outside of Africa, strong selection signals were detected in PF3D7_1475900 (KELT protein, South Asia), PF3D7_1035100 (unknown function, Southeast Asia), and PF3D7_1035300 ( GLURP , Southeast Asia). To complement signals of recent positive selection, we identified and characterised genomic regions enriched for such signals and annotated the associated gene products. A total of 41 unique genomic regions showed enrichment for recent positive selection ( Supplementary Table 11 ), highlighting loci potentially shaped by local adaptation. These regions encoded gene products linked to erythrocyte invasion, drug resistance, virulence, and the regulation of gene expression, including epigenetic modification. Seven regions, East Africa, Central Africa, South Central Africa, Horn of Africa, West Africa, Oceania, and South Asia, shared a selection signal on chromosome 8 (1.300,000–1,330,000), encoding the PHISTc protein family. Another prominent signal was observed on chromosome 4 (1,090,000–1,120,000), present across East Africa, Central Africa, South Central Africa, West Africa, Oceania and South Asia. This region encodes the erythrocyte binding antigen-165 (EBA-165), reinforcing the role of erythrocyte invasion pathways as key targets of selection. Central Africa also exhibited strong positive selection across chromosome 7 (400,000–430,000), which contains pfcrt , which is linked to chloroquine resistance. Additional regions of interest include chromosome 8 (1,350,000–1,380,000), encoding HSP70x , a protein involved in the export of virulence factors, and a region encoding BDP4 , an epigenetic regulator of gene expression observed under selection across multiple regions 42 . Furthermore, the detection of selection around pfubp1 in East Africa is notable given emerging evidence linking this gene to antimalarial resistance 43 . Collectively, these findings highlight chromosomal hotspots where adaptive pressures, particularly those related to host-parasite interactions and drug pressure, have likely shaped P. falciparum evolution. Between-population (XP-EHH) analyses revealed a total of 266 SNPs across 75 genes with evidence of divergent selection between regions (Threshold:>5) ( Supplementary Table 12 ). The greatest number of differentiated sites were observed in comparisons involving Central Africa, particularly with East Africa (n = 152), West Africa (n = 140) and South Central Africa (n = 134). Several genes exhibited high numbers of XP-EHH signals across comparisons, including PF3D7_0713000 (RIF), PF3D7_0709300 (CG2), PF3D7_0710200 , PF3D7_1475900 , and PF3D7_0113800 , each containing over 100 significant sites. GO term analysis of genes with elevated XP-EHH signals identified significant fold changes for biological processes involved in interspecies interactions (P < 0.05; Supplementary Table 13 ). Notably, this included PF3D7_0209000 ( P230 ), which is essential for ookinete formation and mosquito transmission (Central vs. East and South Central Africa), and PF3D7_1346800 ( P47 ), which mediates mosquito immune evasion (East vs. West Africa and Central vs West Africa). These findings suggest that positive selection in different regions may be influenced by local adaptation to Anopheles mosquito vectors. Further investigation of chromosomal regions enriched for XP-EHH signals revealed 39 distinct regions under differential selection across populations ( Supplementary Table 14 ). Of particular interest were two regions associated with antimalarial resistance. These included a locus encoding PPPK-DHPS on chromosome 8 (530,000–550,000), associated with sulfadoxine resistance (Central vs. East or West Africa), and the pfcrt gene on chromosome 7 (400,000–430,000; 360,000–450,000), implicated in chloroquine resistance. XP-EHH signals at the pfcrt locus were detected in comparisons such as Central Africa vs. East Africa, Horn of Africa vs. South Central Africa and South Asia vs. South Central Africa, among others. These findings suggest that variation in historical drug treatment regimens may have driven region-specific selection pressures across P. falciparum populations. Signals of selection vary across countries over time To complement haplotype-based approaches, we also examined signatures of recent positive selection using the iR statistic, which identifies loci with excess IBD sharing. The top five genes with the strongest selection signals (−log₁₀P > 5) for each country and time period are reported ( Supplementary Figure 10, Supplementary Table 15 ). Several genes consistently exhibited strong selection signals across multiple countries and time frames, including ABCK1 , AQP2 , ARO , ATPase2 , CAF1 , CARM1 , CG2 , CRK3 , pfcrt , CUL1 , DHHC4 , DRN1 , JmjC1 , MC-2TM , and members of the FIKK family. Among these, pfcrt and pfdhps ( PPPK-DHPS ), particularly in Papua New Guinea (2012–2015), were notable for their well-established roles in antimalarial drug resistance. In addition to these canonical resistance genes, ABCK1 (an ATP-binding cassette transporter) and MC-2TM (a multidrug and toxin extrusion transporter) also showed strong signals of selection, suggesting possible involvement in drug efflux mechanisms and warranting further functional investigation. Some genes showed evidence of recent positive selection in specific countries only, pointing to localised adaptation. These include ARF-GAP and MCM3 (The Gambia, 2016–2019); CDPK6 , CDPK7 , CK2α , and CLK3 (Kenya, 2016–2019); DHHC9 (Laos, 2016–2019); HAS1 (Benin, 2016–2019); IMC1g (The Gambia, 2016–2019); LSA1 and M712 (Zambia, 2016–2019); NHE (Guinea, 2016–2019); P36 (Papua New Guinea, 2016–2019); PF3D7_0201300 (Mozambique, 2016–2019); and RAB7 (Guyana, 2020–2021). These genes are involved in key cellular processes including metabolism, cell division, signal transduction, protein synthesis, membrane trafficking, and host-cell invasion. Several of the genes under country-specific selection, such as LSA1 (Liver Stage Antigen 1) and P36 , are potential malaria vaccine candidates. Evidence of selection at these loci may have implications for vaccine efficacy and highlights the need to monitor adaptive evolution in P. falciparum populations across geographic regions and over time 44,45 . Deviation from neutral evolution Tajima’s D is a widely used statistic for identifying regions of the genome that deviate from neutrality, offering insights into evolutionary processes such as balancing selection (characterised by an excess of intermediate-frequency variants; Tajima’s D > 2) and selective sweeps (marked by an excess of rare alleles; Tajima’s D < –2) ( Supplementary Figure 11 ). Relatively few 10 kb genomic windows were identified as being under balancing selection across all geographical regions, with the highest numbers observed in South America (n=16), Oceania (n=11), Southeast Asia (n=8), Horn of Africa (n=6), South Asia (n=4), Central Africa (n=2), and South Central Africa (n=1). These regions predominantly map to loci previously implicated in immune-mediated selection. In contrast, a larger number of genomic regions showed evidence of an excess of rare alleles, potentially indicative of recent or partial selective sweeps. The number of such regions varied across geography: Oceania (n=1,938), South America (n=1,797), Horn of Africa (n=1,751), Southeast Asia (n=1,416), South Asia (n=766), South Central Africa (n=337), Central Africa (n=289), East Africa (n=149), and West Africa (n=99). While many of these may represent false positives, as suggested by comparisons with complementary statistics like H12, some may reflect soft selective sweeps that evade detection by metrics such as iHS. A notable example is a region on chromosome 13 (1720000–1820000), which includes the pfkelch13 gene, known to contain markers of artemisinin resistance. This region displayed low Tajima’s D values across multiple populations: East Africa (–2.60), West Africa (–2.56), South Central Africa (–2.54), Central Africa (–2.49), Southeast Asia (–2.28) and South Asia (–2.24). Geographical distribution of drug-resistance markers We first examined the geographical distribution of known genotypic markers associated with antimalarial drug resistance. Established drug resistance markers were obtained from the Malaria-Profiler database and the WHO watchlist for pfkelch13 artemisinin resistance markers 20,46 . The highest combined prevalence of resistance to chloroquine, pyrimethamine, and sulfadoxine was observed in South America (69.5%), followed by South Asia (67.0%), West Africa (62.2%), Central Africa (61.3%), East Africa (57.2%), Oceania (57.1%), South Central Africa (46.3%) and the Horn of Africa (40.3%) ( Fig. 3 ). In contrast, Southeast Asia showed the highest proportion of genotypes resistant to artemisinin in combination with these three drugs (40.1%), reflecting the region’s long-standing issue with multidrug-resistant P. falciparum . Given the global dependence on ACTs as the frontline treatment for P. falciparum malaria, the emergence and spread of artemisinin resistance poses a major threat to malaria control efforts. While artemisinin resistance remains most prevalent in Southeast Asia, genotypic markers were also detected in Oceania (n = 2), South Central Africa (n = 1), and South America (n = 1). The most commonly detected artemisinin resistance–associated mutation was pfkelch13 C580Y, present in 2,270 isolates from Southeast Asia and in two samples from Oceania. Other notable pfkelch13 variants observed in Southeast Asia included P441L (2%), R539T (2%), F446I (1%) and Y493H (1%) ( Supplementary Figure 12 ). Outside Southeast Asia, a single isolate from Zambia (South Central Africa) carried the P441L mutation, which is linked to partial resistance, while the R561H variant, a validated marker of artemisinin resistance, was detected in an isolate from Guyana (South America). A mutation at the same codon as C580Y, C580F, was also identified in one isolate from Myanmar, suggesting continued diversification of resistance-associated alleles in the region. Other mutations associated with full and partial artemisinin resistance have been detected through genotyping and targeted sequencing efforts in countries such as Uganda (R561H, A675V), Rwanda (R561H), Tanzania (C469Y, R561H), Kenya (C469Y, P574L) and Ethiopia (F446I, R662I, P574L); however, these mutations were not observed in the present dataset for these regions 17,47-50 . We further searched for missense mutations in the propeller domain of Kelch13 ( Supplementary Figure 13 ). Fourteen mutations were uniquely observed in West Africa (N458D, E509D, V534L, A557S, T573S, E596G, E612D, G665S, E691D, L722V, C532S, V637I, V566I and V589I), while three mutations were detected in East Africa (Y630F, I634L and S522C). Additionally, the pfkelch13 mutations Q613E and R622T were identified in Central Africa and South Central Africa, respectively. However, all were reported at low frequency (<5 samples total). The N458D mutation is located near N458Y, a validated marker of artemisinin resistance; however, N458D itself is not validated nor widely reported as being associated with resistance 51 . The pfkelch13 A578S variant was observed in Central Africa (n=2), East Africa (n=5), South Asia (n=5), South Central Africa (n=1), and West Africa (n=21), but is not considered linked to drug resistance and is known to be commonly found across Africa 52 . The pfkelch13 S522C mutation has been reported in West Africa (n=4) and East Africa (n=1); although it has been associated with delayed parasite clearance, it has not been designated a candidate marker by WHO due to limited supporting data 52,53 . IBD patterns and SNP associations with pfkelch13 -mediated resistance Our previous analyses did not identify strong signals of positive selection at the pfkelch13 locus specifically; however, the spread of pfkelch13 mutations may involve polygenic adaptation, which can obscure traditional signatures of selective sweeps 54 . To investigate this further, we performed IBD analyses to map transmission networks and characterise the genomic background associated with pfkelch13 mutations, focusing on Southeast Asia and Oceania ( Fig. 4 ). Overall, pfkelch13 C580Y-positive samples exhibited significantly higher pairwise IBD fractions (median: 0.276) compared to samples with other pfkelch13 missense mutations in the propeller domain (median: 0.06; Wilcoxon, P < 2 × 10⁻¹⁶) or wild-type alleles (median: 0.029; Wilcoxon, P < 2 × 10⁻¹⁶) ( Fig. 4 ). These findings support the hypothesis of clonal expansion of the C580Y genotype, particularly in the context of the KEL1/PLA1 lineage, a multidrug-resistant strain associated with piperaquine resistance 55 . To gain further insight, IBD networks were constructed using thresholds of 47.5% and 95% IBD ( Supplementary Figures 14–16 ). These thresholds were chosen to reflect broader regional transmission (47.5%) and more direct, local transmission (95%). While the convenience sampling approach may limit the ability to fully reconstruct historical transmission patterns, data were stratified into three epidemiologically relevant time windows: 2008–2011 (emergence of pfkelch13 C580Y in western Cambodia), 2012–2015 (expansion of artemisinin resistance and the rise of KEL1/PLA1) and 2016–2019 (regional dominance of artemisinin resistance across Southeast Asia) ( Supplementary Figures 14–16 ). At the 95% IBD threshold, samples consistently clustered by pfkelch13 genotype across all time periods, indicating transmission of closely related parasites carrying the same resistance mutations. Clustering also occurred by country or site, suggesting either local transmission or the onward spread of imported cases. In contrast, at the 47.5% threshold, clusters frequently included a mix of pfkelch13 genotypes, likely reflecting recombination over time among genetically related parasites sharing similar genomic backgrounds. One example is a large cluster from 2008–2011 containing three genotypes ( pfkelch13 C580Y, pfkelch13 539T, and wild-type) across Cambodia, Vietnam and Laos ( Supplementary Figure 14 ). Over time, both the number and geographic spread of pfkelch13 C580Y-positive clusters increased, often spanning multiple countries. This trend likely reflects the expansion and regional dominance of the KEL1/PLA1 lineage, although definitive conclusions are limited by the convenience sampling design. The IBD analysis revealed clonal clusters of pfkelch13 C580Y-positive parasites with extensive shared genomic regions, suggesting that additional loci may be under co-selection. To investigate this further, we performed genome-wide SNP association analyses that integrated co-occurrence patterns, genotypic correlations, and population-adjusted models to identify variants linked to the C580Y background across Southeast Asia (see Methods ). Candidate mutations were prioritised into three confidence tiers: Tier 1 (high), Tier 2 (moderate) and Tier 3 (low, functional relevance-based) (see Methods ). A total of 22 mutations across 20 genes met the Tier 1 criteria, showing strong support across multiple analyses, significant association (P 0.5), and large effect sizes (an odds ratio (OR) in the 95th percentile) (Supplementary Table 16 ). This set included ARPS10 V127M, a previously reported interactor with pfkelch13 C580Y, validating our approach 56 . Another top-ranked variant, MyoF S969P (PF3D7_1226000, formerly MyoC ), a component of the K13-associated protein complex, further implicates changes in the parasite’s intracellular endocytosis architecture in facilitating resistance in the Greater Mekong sub-region. An additional 55 mutations across 53 genes were classified as Tier 2 (moderate confidence), including pfcrt , which has been linked to the pfkelch13 C580Y background. This supports the idea that a pre-existing drug resistance background, such as chloroquine resistance, may have predisposed parasites to evolve artemisinin resistance. However, recent studies of Ugandan isolates with K13-mediated reduced artemisinin susceptibility indicate that these have evolved on a background of wild-type pfcrt and complete susceptibility to chloroquine 21,22 . Tier 3 consisted of 20 mutations in 14 genes ( MDR2, DHFR-TS, UBP1, AP2-L, MDR1, PF3D7_0907200, MCA2, AP2-G, VPS51, CRT, MyoC/MyoF, PF3D7_1365800, PF3D7_0907200, PF3D7_1329500 , and PF3D7_1243400 ) ( Supplementary Table 16 ). These mutations show weaker statistical associations but were retained due to their potential functional relevance based on prior studies. To investigate potential shared pathways, we constructed a protein–protein interaction (PPI) network using all prioritised candidates ( Fig. 4F ). Six clusters emerged, including one enriched for known drug resistance proteins (MDR1, MDR2, CRT, DHFR-TS) and associated genes ( PF3D7_0214600, PF3D7_0104300, PF3D7_1438500, OXA1, PF3D7_1017000, PF3D7_1331300 ). This suggests that PF3D7_0214600 may have contributed to the emergence of pfkelch13 C580Y. Additional interactions were observed between pfkelch13 and PF3D7_0303800, a Tier 2 candidate found in Laos, as well as among PF3D7_0405400 (putative pre-mRNA processing-splicing factor 8), PF3D7_1119300 (splicing factor U2AF small subunit), and a duplicate listing of PF3D7_1119300 , all of which are involved in mRNA splicing, potentially influencing gene regulation and parasite fitness. Together, these associations highlight a complex genomic background linked to pfkelch13 C580Y. While not all identified mutations may be functionally relevant, some are likely to represent compensatory changes or modifiers that impact cellular pathways, gene expression or survival. Further functional validation is needed to delineate their roles in compensatory evolution and artemisinin resistance. DISCUSSION With the increasing threat of P. falciparum across the globe, substantial effort is required to achieve malaria elimination. Data-driven strategies and the development of new tools offer innovative approaches to aid the surveillance of P. falciparum . However, generating such tools necessitates a comprehensive understanding of P. falciparum on a global scale, which has recently become attainable through access to large-scale, publicly accessible data 26,31-35,37 . By delving further into the P. falciparum genome, we can gain greater insight into the biological mechanisms that threaten malaria control and enhance our understanding of population genetics. This large-scale genomic data allows for a deeper exploration of P. falciparum 's evolution, patterns of gene flow, and the impact of control efforts, offering critical information to guide malaria control strategies. The evolution of P. falciparum has been dynamic, shaped significantly by human and vector migration, which has facilitated its global spread, and by control efforts, which have led to population reductions and bottlenecks 8 . Both historical migrations and recent interventions have left their marks, visible in the population structure analysis of P. falciparum genome-wide SNPs 9 . P. falciparum isolates cluster according to continental boundaries, with some clustering according to specific geographical regions such as South Asia, Southeast Asia, and Oceania, while African samples are less distinct. The clustering of South American samples with African regions aligns with evidence suggesting multiple introductions of P. falciparum from Africa, including during the transatlantic slave trade 36,57 . ADMIXTURE analysis further highlights the shared ancestry between intercontinental and intracontinental regions. For example, as mentioned, South American samples share ancestry with West African populations 57 . However, they also exhibit a distinct genetic signature, likely shaped by a population bottleneck during introduction and approximately 450 years of subsequent isolated evolution, local adaptation, or influence from ancestral lineages not represented in current West African samples. Consistent with multidimensional scaling analysis, African populations exhibited overlapping ancestry, with samples from East and Central Africa showing greater genetic similarity to each other than to those from West Africa. In contrast, the Horn of Africa forms a distinct genetic cluster, shaped by gene flow from African regions as well as some influence from Asia across the Indian Ocean and Arabian Sea. Outside Africa, there was partial overlap in ancestry between Southeast Asia, Oceania and South Asia, reflecting shared evolutionary history or gene flow, although each region retained distinct genetic signatures, indicative of region-specific selection or demographic events. Further exploration of the relationship between genetic and geographic structure was carried out using F ST analysis. Generally, genetic differentiation correlated with geographic distance, except in regions with smaller sample sizes, for which genetic differentiation was only evident at intracontinental scale. Despite the high SNP diversity in African samples, they displayed the least genetic differentiation. This aligns with other analyses indicating a high degree of outcrossing within African populations, supported by higher MOI and lower IBD estimates. Genomic regions with high F ST values provide insight into the drivers of genetic differentiation between geographical regions. Several genes with high F ST sites are known to play roles in transmission, including the gene encoding P47, a protein used for mosquito immune evasion, as well as P48/P45, which is essential for ookinete formation and transmission 58-60 . These genes and their products are proposed targets for transmission-blocking vaccines. Moreover, highly differentiating sites were detected in drug-resistance genes, reflecting the administration of different drugs across various regions and time points. The widespread prevalence of drug-resistance mutations across all regions is concerning and underscores the importance of global surveillance. Sites with high F ST may be utilised for geographical classification and molecular barcoding to support such surveillance efforts. Despite the dataset being curated from samples sequenced over several decades, the results from the population genetic analysis align broadly with current trends in malaria transmission and disease burden. High MOI, indicated by lower F WS values, was estimated for isolates from South Asia, West Africa, East Africa and Central Africa. This corresponds to the high transmission intensity and frequent outcrossing known to occur in these regions and recent global health reports 1,61,62 . Country-level analysis suggested increasing MOI in Mali and Kenya in recent years (2016 onwards), suggesting the sites sampled from these countries should be of key concern and targets for malaria intervention. Conversely, low MOI was reported in the Horn of Africa, South America, Oceania and Southeast Asia, which could indicate less outbreeding and lower transmission because of more effective malaria control measures. However, this may also be influenced by the smaller sample sizes from some of these regions. Specific countries might also exhibit low MOI due to geographic isolation, as previously observed in samples from the Bijagós islands 32 . These findings suggest that P. falciparum infections in South Asia and Southeast Asia should be considered independently during global assessments rather than being combined. This contrasts with observations made for P. vivax whereby F WS values are marginally higher in South Asia compared to Southeast Asia, but the values are overall very similar, likely due to P. vivax relapse maintaining genetic diversity across regions with comparable transmission 63 . The MOI results are consistent with IBD analysis, where higher fractions were observed in the Horn of Africa, South America, Oceania and Southeast Asia, indicative of reduced outcrossing and lower transmission. High IBD fractions were particularly noted in genes involved in drug resistance. This is likely due to strong positive selection from intensive antimalarial drug administration. Under such selective pressure, IBD sharing at drug-resistance loci may increase, alongside neutral genomic regions linked to these genes 64 . This could explain the higher IBD fractions found near pfcrt, which contains chloroquine-resistance mutations. High fractions of IBD were observed in pfkelch13 interaction candidates KIC7 and KIC9 across the Horn of Africa and South America, regions that may be affected by artemisinin-resistance 65 . This could indicate selection in these genes which should consequently be a target for future investigation into markers for artemisinin-resistance. In addition, it has been suggested that regions under strong positive selection can bias IBD analysis 64 . Further research is needed to mitigate this effect for downstream IBD-based inference, such as estimating effective population size. Moreover, the impact of positive selection was evident through the assessment of regions with extended haplotype homozygosity within (iHS) and between (XP-EHH) populations. Across all geographical regions, SNPs in genes encoding cell surface proteins or those involved in host cell invasion, such as the SURFIN and CLAG families, exhibited strong signatures of positive selection. The strong signals of positive selection in SURFIN and CLAG genes likely reflect recent or ongoing adaptive changes to overcome host immunity or improve invasion efficiency, which may coexist with balancing selection maintaining diversity at other sites or times 66,67 . When comparing selection between populations, mosquito interaction genes P47 and P230 were found to have undergone cross-population selection, between regions and countries 58 . This could be underpinned by ecological dynamics. For example, in Southeast Asia, where multiple Anopheles species coexist and the dominant species can vary 68,69 . Signals of positive selection were also detected in drug-resistance genes, such as pfcrt and PPPK-DHPS . This was most likely driven by different treatment regimens implemented across regions at different time points. The role of loci involved in gene expression, such as BDP4 , should also not be ignored. The regulation of gene expression and epigenetic modification support the parasite’s complex lifecycle and provide an additional mechanism for adaptation 70 . Together, examining regions under positive selection offers valuable insights into the past, present and future evolution of P. falciparum and can guide its control. After detecting the influence of anti-malarial treatment across P. falciparum genomes, we aimed to explore the patterns of drug-resistance mutations further. The geographic distribution of these mutations aligned with past and current drug administration programs in each country and region. For instance, pfkelch13 mutations were predominantly observed in samples from Southeast Asia and Oceania 17 . However, these mutations were also observed in South America and South Central Africa, demonstrating the greater need for surveillance of pfkelch13 markers outside of Southeast Asia 18 . Profiling drug-resistant strains could be improved by considering drug-resistant haplotypes and exploring potential compensatory effects further, as well as examining structural variants and copy number variants that were not covered in this study 26 . Using a convenience sample approach is beneficial for examining large-scale drug-resistance data, but caution is necessary when interpreting results due to temporal changes that may occur. The lack of strong positive selection detected in pfkelch13 has indicated that there may be compensatory mechanisms driving the evolution of artemisinin resistance across Southeast Asia 56,71 . We highlight the clonal expansion of pfkelch13 C580Y mutants using IBD analysis, while also showing that other mutations have independently arisen from similar genomic backgrounds. In the absence of phenotypic data, potential interactors with the most common pfkelch13 mutation (C580Y) were probed by leveraging the power of the large-scale genome-wide sequencing data to search for SNP linkage, co-occurrence and association. Some mutations may be ‘background mutations’ associated with the genetic lineage of samples from Southeast Asia, which could be clarified through genome-wide association tests with the artemisinin drug susceptibility phenotype and the inclusion of population structure covariates 72 . Despite this, combined with knowledge of protein-protein interactions from the literature, several candidates were shortlisted as potential candidates, including pfcrt and ARPS10 which have previously been reported 56 , as well as pfubp1, which was reported as undergoing positive selection in East Africa in this study. Additional candidates include MyoF and RAD5. MyoF is associated with the K13 compartment; RAD5 is also reported as being under selection in the Greater Mekong Sub-region, where the KEL1/PLA1 lineage is known to dominate 73,74 . While further validation is needed, the identification of new candidate mutations provides valuable insights into the potential drivers of artemisinin resistance that may emerge in other regions globally. The majority of isolates in the curated dataset were collected before the significant emergence of African pfkelch13 propeller domain variants, which were first reported around 2020 17,47 . As such, further analysis of these variants is limited in this dataset and represents an important area for future research. This study has identified specific patterns of genetic variation in P. falciparum across the globe which varied over time and between countries and regions. Such information, integrated with further validation and additional study can help to rapidly channel genomic surveillance efforts into prompt interventions, with particular focus on screening drug-resistant mutations, adaptation to the Anopheles vector and geographical classification 20 . METHODS Sequence Data and Pre-processing raw reads A combined dataset comprised of 23,462 P. falciparum WGS were considered for analysis, including 16,203 high quality samples which were obtained from the MalariaGen Pf7 data resource 26 , 7,183 previously published samples 31-35,37,64,75,76 and 76 newly sequenced isolates (Brazil and Vietnam) which were processed for this study using previously described methods 31 . After pre-processing the raw genomic data, duplicate samples, mixed species and low-quality samples were removed from the dataset, leaving 17,565 high quality samples for further analysis ( Supplementary Table 17 ). Raw paired end sequence data was processed using an established bioinformatic pipeline (https://github.com/LSHTMPathogenSeqLab/fastq2matrix). FastQ files were obtained from the European Nucleotide Archive and were trimmed to remove poor quality sequences using trimmomatic(v0.39) and the following parameters: LEADING:3 TRAILING:3 SLIDINGWINDOW:4:20 MINLEN:36 77 . Trimmed sequences were subsequently aligned to the P. falciparum 3D7 reference genome v3 (GCA_000002765.3) using bwa-mem (v0.7.17-r1188) to produce BAM files 78 . Samtools v1.18 (fixmate and markdup) was used to correct mate information after mapping with bwa-mem 79 . Base quality score recalibration (BQSR) and correction were then performed using GATK (v4.1.4.1) (BaseRecalibrator and ApplyBQSR) to reduce systematic errors in the quality score of base calls derived from the sequencing process 80 . This was carried out using the P. falciparum genetic crosses 1.0 dataset (https://www.malariagen.net/data_package/pf-crosses-1-0/). Variant calling (SNPs and small Indels) were called by also using GATK (v4.1.4.1) software (HaplotypeCaller) to create per-sample gVCF files (parameters: -ERC GVCF) 81 . The GenomicsDB datastore was used to store validated VCFs via the GATK’s GenomicsDBImport function. A multi-sample VCF file was then created using the GenotypeGVCFs function and further quality score recalibration was carried out using the GATK Variant Quality Score Recalibration (VQSR) function (parameters: -an QD -an FS -an SOR -an DP -maxGaussians 8 and -mq-cap-for-logit-jitter-transform 70). Variant Quality Score Log-Odds (VQSLOD) were obtained using the ApplyVQSR function (parameter: -truth-sensitivity-filter-level 99.0, P. falciparum genetic crosses 1.0 dataset) and variants with VQSLOD score < 0 were filtered out to retain high-quality variant calls. Only SNPs found within the core genome were retained in the dataset. Further quality metrics were used to ensure the remaining dataset was of high quality. Isolates with >40% missing data were removed, leaving high quality isolates for population genetic analysis that had consistent coverage across the core genome. The genotype of SNPs with mixed calls with a secondary MAF>20% were determined by the ratio of coverage. Variants were annotated using snpEff (v5.1). After all filtering steps, 17,565 high quality samples remained for further analysis. The high-quality multi-sample VCF was filtered for bi-allelic SNPs. Two VCF files were generated: one containing only bi-allelic SNPs, and a second containing normalized multi-allelic sites, where only the alternative allele with the highest minor allele frequency (MAF) was retained. The normalised VCF underwent additional filtering to remove additional hypervariable genes, which were defined as being in the top 95% quartile for SNP density (SNP per bp) ( Supplementary Table 18 ). Population Genetic Analysis Multidimensional scaling (MDS) was carried out using all samples and SNPs. A distance matrix was first calculated using PLINK (v1.90) software using a filtered, bi-allelic VCF file 82 . The MDS was carried out over the distance matrix using R (v4.2.2). F ST analysis and estimation of SNP diversity (π) was carried out using the python package scikit-allele (v1.37) (https://github.com/cggh/scikit-allel). The VCF file was converted to a ZARR format and F ST analysis was performed between regions, as well as a ‘one-against-all’ approach whereby samples from each region were compared to the rest of the dataset in the analysis. F ST analysis was only carried out using segregating sites. Thresholds of F ST >0.75 (moderate) and F ST >0.95 (high) were used to identify sites with moderate-high F ST . Correlation between genetic and geographic distances (F ST ) were estimated using the Mantel test using the vegan package in R (https://github.com/vegandevs/vegan). Mean pairwise difference was used to estimate SNP diversity in 10kb windows, excluding sites with >80% missingness and SNPs in non-coding regions. However, results from SNP diversity analyses are heavily influenced by missingness thresholds and lack of invariant sites so may not be reliable estimates and should be interpreted with caution. All code for these analyses is available on GitHub in a dedicated repository: https://github.com/NinaMercedes/PopGen/tree/main/Pop_Gen. Multiplicity of infection (MOI) was first estimated using only coding regions and bi-allelic SNPs. A population-specific MAF threshold of 1% was used to filter SNPs (threshold for all analyses unless stated otherwise). The multi-sample VCFs were split and filtered according to country and region using bcftools (v1.20) and the F WS metric was calculated using the moimix R package (v0.0.2.9001) (https://github.com/bahlolab/moimix). A F WS metric of >0.95 was used as a threshold to remove multiclonal isolates. F WS scores were compared across countries and year groups to identify changes F WS score using ANOVA. Year groups were defined on a distribution basis and spanned 3-year periods. P-values were adjusted using the false discovery rate and P<0.005 was used to indicate a significant difference between groups. Downstream analysis was performed using biallelic SNPs only which were used to produce a binary genotype matrix (reference allele=0, alternative allele=1, mixed allele=0.5, missing allele=Ns). ADMIXTURE software (v1.3) (parameters: -cv=10 -j8 --haploid="*") was used to estimate individual ancestries across SNP genotypes 83 . This was performed using a bed file which was converted from the multi-sample biallelic VCF file using PLINK (v1.90). Ten-fold cross-validation was used to estimate the optimal K value through inspection of the inflection point after running across K values 1 to 12. The optimal value was K=10 ( Supplementary Figure 17 ). Furthermore, IBD sharing across genomes was estimated using hmmIBD software which utilised a hidden Markov chain model 84 . The pairwise fraction collected for all specified regions and countries was used to estimate genomic relatedness. Genome-wide IBD fractions across 10kb sliding windows were also calculated. Regions were annotated with gene annotations and gene products for enhanced interpretation. Analysis of IBD at country and year group level were also performed using isoRelate and the iR statistic was used to identify sites undergoing positive selection 85 . The biallelic matrix was also used to determine regions under positive selection within integrated haplotype homozygosity score (Threshold: −log10[1−2|Φ(iHS)-0.5|]>4.0) and between populations (XP-EHH) (Threshold: 5) using the rehh R package 86 . Highly variable regions such as var genes were removed from the analysis to prevent false positive results. The iHS is a within-population statistic, where a positive score indicates favourable selection for the ancestral allele and negative score indicates favourable selection for the alternative allele. The direction of the XP-EHH output corresponds to the population in which selection is taking place. Regions under positive selection were also annotated with gene annotations and gene products. Countries with fewer than 5 samples were excluded from population genetic analyses were applicable. Tajima’s D was estimated using VCFtools (v0.1.16) in 10kb windows. Genes of interest identified across regional level F ST analyses (single and paired), IBD analysis, iHS and XPEHH, were annotated using GO terms (biological process, molecular function and cellular components) and KEGG pathways using PANTHER 87 . All selection analyses were also run using the normalised VCF file and binary matrix outputs. This was to prevent the exclusion of any important SNPs, including pfkelch13 C580Yin selection analyses. However, this did not have a large overall impact on the overall results, except for the inclusion of additional variation. Drug-resistance mutations and pfkelch13 C580Y associations Drug-resistance mutations were assessed using the multi-allelic VCF file. Known drug-resistance mutations were retrieved from the malaria-profiler database (https://github.com/jodyphelan/malaria-db) and https://www.who.int/news-room/questions-and-answers/item/artemisinin-resistance which were used to filter the VCF files 20 . Additional nonsynonymous mutations were used by filtering gene boundaries using a bed file found within the database. This was used to generate a matrix for identifying additional mutations and nonsynonymous mutation combinations. Indels were also included in the matrix (PF3D7_0523000 N75E). Samples from Southeast Asia were subject to further IBD analysis. Outputs from hmmIBD were used to construct identity-by-descent (IBD) networks across Southeast Asia, across three-year time periods spanning from 2008 to 2019. Networks were visualised using ipysigma (https://github.com/medialab/ipysigma). 47.5% and 95% IBD thresholds were used to reveal broader, ancestral connectivity and recent, direct clonal expansion. Pairwise IBD fractions were compared across three genotypic groups: pfkelch13 C580Y, other missense mutations in the propeller domain, and wild-type. Statistical significance was assessed using the Kruskal-Wallis test for overall group differences, followed by two-sided Mann-Whitney U tests for pairwise comparisons. A multi-layered association analysis was conducted to identify SNPs associated with the pfkelch13 C580Y mutation using both a drug-resistance binary matrix and a genome-wide binary matrix filtered for common variants (MAF > 1%). Analyses were performed using two statistical frameworks: a logistic regression model adjusted for the top five principal components (across all samples, “All”), and latent factor models within each country (Cambodia, Laos, Vietnam, Myanmar, and Thailand), implemented in the LEA package to account for population structure and admixture (optimal K inferred via snmf) 88 . In both cases, P-values were adjusted using the Benjamini-Hochberg method. To complement the association models, we calculated an interaction coefficient (odds ratio, OR) was calculate, where OR = A×D / B×C based on co-occurrence of SNPs with C580Y (0 counts imputed as 0.5), and genotypic correlation ( r² ) using VCFtools (v0.1.16). These four metrics, adjusted p-value from the logistic model, adjusted p-value from latent factor models, OR, and r² were integrated to prioritise SNPs into low, medium and high confidence candidates (‘Tiers’). Tier 1: SNPs meeting at least three of the four criteria (Adjusted P 0.5, OR > 123). Tier 2: SNPs meeting two criteria, with moderate support (r² > 0.2 and OR > 5). Tier 3: SNPs meeting one criterion and located in a curated list of genes with biological relevance or prior evidence ( Supplementary Table 19 ) 74 . STRING protein-protein interactions (clustered using the k-means algorithm) was used to identify functional links between gene candidates 89 . Declarations AUTHORS CONTRIBUTIONS SC and TGC conceived and supervised the project. JGD, LV, SM, JG, CJS, CRFM, NTHN, NTHB, and NQT conducted sample processing and DNA extraction. JGD, LV, SM, and SC conducted sequencing. JT provided software tools. JGD, LV, SM, JG, CJS, CRFM, NTHN, NTHB, and NQT provided data. NB performed bioinformatics and statistical analyses under the supervision of SC and TGC. All authors contributed to data interpretation. NB drafted the manuscript, with input from all authors. All authors reviewed, edited, and approved the final manuscript. NB, SC, and TGC compiled the final submission. DECLARATION OF INTERESTS Authors declare no competing interests. DATA SHARING Raw sequencing data are available in the European Nucleotide Archive (ENA). A complete list of accession numbers is provided in PRJEB94034. ACKNOWLEDGEMENTS The study was funded by MRC Newton UK – MOST Vietnam (ref. MR/R026297/1) award. CRFM was supported by the São Paulo Research Foundation-FAPESP (www.fapesp.br) [grant numbers 2024/10186-9] and the National Council for Scientific and Technological Development-CNPq (www.cnpq.br) [grant number 307193/2023-3]. JGD was supported by fellowships from FAPESP [grant numbers 2019/12068-5 and 2022/02771-3]. TGC and SC are funded by the UKRI (BBSRC BB/X018156/1; MRC MR/X005895/1; EPSRC EP/Y018842/1). The funders had no role in study design, data collection and analysis, decision to publish, or the preparation of the manuscript. References Geneva World Health Organization. World malaria report 2024: addressing inequity in the global malaria response. Geneva: World Health Organization; 2024. Licence: CC BY-NC-SA 3.0 IGO. (2024). Masserey, T. et al. The influence of biological, epidemiological, and treatment factors on the establishment and spread of drug-resistant Plasmodium falciparum. eLife 11 , e77634 (2022). https://doi.org/10.7554/eLife.77634 Adams, K. L. et al. Selection for insecticide resistance can promote Plasmodium falciparum infection in Anopheles. PLOS Pathogens 19 , e1011448 (2023). https://doi.org/10.1371/journal.ppat.1011448 Leal Filho, W., May, J., May, M. & Nagy, G. J. Climate change and malaria: some recent trends of malaria incidence rates and average annual temperature in selected sub-Saharan African countries from 2000 to 2018. Malaria Journal 22 , 248 (2023). https://doi.org/10.1186/s12936-023-04682-4 Yalcindag, E. et al. Multiple independent introductions of Plasmodium falciparum in South America. Proc Natl Acad Sci U S A 109 , 511-516 (2012). https://doi.org/10.1073/pnas.1119058109 Tanabe, K. et al. Plasmodium falciparum accompanied the human expansion out of Africa. Curr Biol 20 , 1283-1289 (2010). https://doi.org/10.1016/j.cub.2010.05.053 Tanabe, K. et al. Plasmodium falciparum mitochondrial genetic diversity exhibits isolation-by-distance patterns supporting a sub-Saharan African origin. Mitochondrion 13 , 630-636 (2013). https://doi.org/10.1016/j.mito.2013.08.008 Tagliamonte, M. S. et al. Genetic Markers of Adaptation of Plasmodium falciparum to Transmission by American Vectors Identified in the Genomes of Parasites from Haiti and South America. mSphere 5 (2020). https://doi.org/10.1128/mSphere.00937-20 Rodrigues, P. T. et al. Human migration and the spread of malaria parasites to the New World. Sci Rep 8 , 1993 (2018). https://doi.org/10.1038/s41598-018-19554-0 Anderson, T. J. et al. Microsatellite markers reveal a spectrum of population structures in the malaria parasite Plasmodium falciparum. Mol Biol Evol 17 , 1467-1482 (2000). https://doi.org/10.1093/oxfordjournals.molbev.a026247 Machado, R. L. et al. Genetic structure of Plasmodium falciparum populations in the Brazilian Amazon region. J Infect Dis 190 , 1547-1555 (2004). https://doi.org/10.1086/424601 Iwagami, M. et al. Genetic diversity and population structure of Plasmodium falciparum in the Philippines. Malar J 8 , 96 (2009). https://doi.org/10.1186/1475-2875-8-96 Mobegi, V. A. et al. Population genetic structure of Plasmodium falciparum across a region of diverse endemicity in West Africa. Malar J 11 , 223 (2012). https://doi.org/10.1186/1475-2875-11-223 Sirawaraporn, W., Sathitkul, T., Sirawaraporn, R., Yuthavong, Y. & Santi, D. V. Antifolate-resistant mutants of Plasmodium falciparum dihydrofolate reductase. Proc Natl Acad Sci U S A 94 , 1124-1129 (1997). https://doi.org/10.1073/pnas.94.4.1124 Staedke, S. G. et al. Relationship between age, molecular markers, and response to sulphadoxine-pyrimethamine treatment in Kampala, Uganda. Trop Med Int Health 9 , 624-629 (2004). https://doi.org/10.1111/j.1365-3156.2004.01239.x Kublin, J. G. et al. Molecular markers for failure of sulfadoxine-pyrimethamine and chlorproguanil-dapsone treatment of Plasmodium falciparum malaria. J Infect Dis 185 , 380-388 (2002). https://doi.org/10.1086/338566 Uwimana, A. et al. Emergence and clonal expansion of in vitro artemisinin-resistant Plasmodium falciparum kelch13 R561H mutant parasites in Rwanda. Nat Med 26 , 1602-1608 (2020). https://doi.org/10.1038/s41591-020-1005-2 Njiro, B. J. et al. Molecular surveillance of chloroquine-resistant Plasmodium falciparum in sub-Saharan African countries after withdrawal of chloroquine for treatment of uncomplicated malaria: A systematic review. J Infect Public Health 15 , 550-557 (2022). https://doi.org/10.1016/j.jiph.2022.03.015 Ravenhall, M. et al. An analysis of large structural variation in global Plasmodium falciparum isolates identifies a novel duplication of the chloroquine resistance associated gene. Scientific Reports 9 , 8287 (2019). https://doi.org/10.1038/s41598-019-44599-0 Phelan, J. E. et al. Rapid profiling of Plasmodium parasites from genome sequences to assist malaria control. Genome Med 15 , 96 (2023). https://doi.org/10.1186/s13073-023-01247-7 van Schalkwyk, D. A. et al. Treatment Failure in a UK Malaria Patient Harboring Genetically Variant Plasmodium falciparum From Uganda With Reduced In Vitro Susceptibility to Artemisinin and Lumefantrine. Clin Infect Dis 78 , 445-452 (2024). https://doi.org/10.1093/cid/ciad724 Tumwebaze, P. K. et al. Decreased susceptibility of Plasmodium falciparum to both dihydroartemisinin and lumefantrine in northern Uganda. Nat Commun 13 , 6353 (2022). https://doi.org/10.1038/s41467-022-33873-x Volkman, S. K. et al. A genome-wide map of diversity in Plasmodium falciparum. Nature Genetics 39 , 113-119 (2007). https://doi.org/10.1038/ng1930 Sabeti, P. C. et al. Detecting recent positive selection in the human genome from haplotype structure. Nature 419 , 832-837 (2002). https://doi.org/10.1038/nature01140 Neafsey, D. E., Taylor, A. R. & MacInnis, B. L. Advances and opportunities in malaria population genomics. Nat Rev Genet 22 , 502-517 (2021). https://doi.org/10.1038/s41576-021-00349-5 Abdel Hamid, M. M. et al. Pf7: an open dataset of Plasmodium falciparum genome variation in 20,000 worldwide samples. Wellcome Open Res 8 , 22 (2023). https://doi.org/10.12688/wellcomeopenres.18681.1 Abera, D. et al. Genomic analysis reveals independent evolution of Plasmodium falciparum populations in Ethiopia. Malar J 20 , 129 (2021). https://doi.org/10.1186/s12936-021-03660-y Mwesigwa, A. et al. Genetic diversity and population structure of Plasmodium falciparum across areas of varied malaria transmission intensities in Uganda. Malar J 24 , 97 (2025). https://doi.org/10.1186/s12936-025-05325-6 Coulibaly, A. et al. Genome-wide SNP analysis of Plasmodium falciparum shows differentiation at drug-resistance-associated loci among malaria transmission settings in southern Mali. Front Genet 13 , 943445 (2022). https://doi.org/10.3389/fgene.2022.943445 Neafsey Daniel, E. et al. Genetic Diversity and Protective Efficacy of the RTS,S/AS01 Malaria Vaccine. New England Journal of Medicine 373 , 2025-2037 (2015). https://doi.org/10.1056/NEJMoa1505819 Vanheer, L. N. et al. Genome-wide genetic variation and molecular surveillance of drug resistance in Plasmodium falciparum isolates from asymptomatic individuals in Ouélessébougou, Mali. Sci Rep 13 , 9522 (2023). https://doi.org/10.1038/s41598-023-36002-w Moss, S. et al. Population dynamics and drug resistance mutations in Plasmodium falciparum on the Bijagós Archipelago, Guinea-Bissau. Sci Rep 13 , 6311 (2023). https://doi.org/10.1038/s41598-023-33176-1 Osborne, A. et al. Plasmodium falciparum population dynamics in East Africa and genomic surveillance along the Kenya-Uganda border. Sci Rep 14 , 18051 (2024). https://doi.org/10.1038/s41598-024-67623-4 Fola, A. A. et al. Genomics reveals heterogeneous Plasmodium falciparum transmission and selection signals in Zambia. Communications Medicine 4 , 67 (2024). https://doi.org/10.1038/s43856-024-00498-8 Florimond, C. et al. Impact of piperaquine resistance in Plasmodium falciparum on malaria treatment effectiveness in The Guianas: a descriptive epidemiological study. The Lancet Infectious Diseases 24 , 161-171 (2024). https://doi.org/https://doi.org/10.1016/S1473-3099(23)00502-9 Lefebvre, M. J. M. et al. Population Genomic Evidence of Adaptive Response during the Invasion History of Plasmodium falciparum in the Americas. Molecular Biology and Evolution 40 , msad082 (2023). https://doi.org/10.1093/molbev/msad082 Vanhove, M. et al. Temporal and spatial dynamics of Plasmodium falciparum clonal lineages in Guyana. PLoS Pathog 20 , e1012013 (2024). https://doi.org/10.1371/journal.ppat.1012013 Otto, T. D. et al. Genomes of all known members of a Plasmodium subgenus reveal paths to virulent human malaria. Nature Microbiology 3 , 687-697 (2018). https://doi.org/10.1038/s41564-018-0162-2 Bopp, S. et al. Potent acyl-CoA synthetase 10 inhibitors kill Plasmodium falciparum by disrupting triglyceride formation. Nature Communications 14 , 1455 (2023). https://doi.org/10.1038/s41467-023-36921-2 Keleta, Y., Ramelow, J., Cui, L. & Li, J. Molecular interactions between parasite and mosquito during midgut invasion as targets to block malaria transmission. npj Vaccines 6 , 140 (2021). https://doi.org/10.1038/s41541-021-00401-9 Kats, L. M. et al. An exported kinase (FIKK4.2) that mediates virulence-associated changes in Plasmodium falciparum-infected red blood cells. Int J Parasitol 44 , 319-328 (2014). https://doi.org/10.1016/j.ijpara.2014.01.003 Day, J., Passecker, A., Beck, H. P. & Vakonakis, I. The Plasmodium falciparum Hsp70-x chaperone assists the heat stress response of the malaria parasite. Faseb j 33 , 14611-14624 (2019). https://doi.org/10.1096/fj.201901741R Xu, R. et al. Deaggregation of mutant Plasmodium yoelii de-ubiquitinase UBP1 alters MDR1 localization to confer multidrug resistance. Nature Communications 15 , 1774 (2024). https://doi.org/10.1038/s41467-024-46006-3 Cummings, J. F. et al. Recombinant Liver Stage Antigen-1 (LSA-1) formulated with AS01 or AS02 is safe, elicits high titer antibody and induces IFN-gamma/IL-2 CD4+ T cells but does not protect against experimental Plasmodium falciparum infection. Vaccine 28 , 5135-5144 (2010). https://doi.org/10.1016/j.vaccine.2009.08.046 Yadav, N. et al. Identifying Plasmodium P36 and P52 antigens for coadministration with circumsporozoite protein to enhance vaccine efficacy. npj Vaccines 9 , 241 (2024). https://doi.org/10.1038/s41541-024-01040-6 World Health Organization. Malaria: Artemisinin partial resistance. https://www.who.int/news-room/questions-and-answers/item/artemisinin-resistance. (2025). Balikagala, B. et al. Evidence of Artemisinin-Resistant Malaria in Africa. N Engl J Med 385 , 1163-1171 (2021). https://doi.org/10.1056/NEJMoa2101746 Juliano, J. J. et al. Prevalence of mutations associated with artemisinin partial resistance and sulfadoxine-pyrimethamine resistance in 13 regions in Tanzania in 2021: a cross-sectional survey. Lancet Microbe 5 , 100920 (2024). https://doi.org/10.1016/s2666-5247(24)00160-5 Osoti, V. et al. Detection of low frequency artemisinin resistance mutations, C469Y, P553L and A675V, and fixed antifolate resistance mutations in asymptomatic primary school children in Kenya. BMC Infectious Diseases 25 , 73 (2025). https://doi.org/10.1186/s12879-025-10462-z Alemayehu, A. A. et al. Expansion of the Plasmodium falciparum Kelch13 R622I Mutation in Northwest Ethiopia. Open Forum Infectious Diseases 12 , ofaf279 (2025). https://doi.org/10.1093/ofid/ofaf279 Tornyigah, B. et al. Effect of Drug Pressure on Promoting the Emergence of Antimalarial-Resistant Parasites among Pregnant Women in Ghana. Antimicrobial Agents and Chemotherapy 64 , 10.1128/aac.02029-02019 (2020). https://doi.org/10.1128/aac.02029-19 Schmedes, S. E. et al. Plasmodium falciparum kelch 13 Mutations, 9 Countries in Africa, 2014-2018. Emerg Infect Dis 27 , 1902-1908 (2021). https://doi.org/10.3201/eid2707.203230 World Health Organization. World Health Organization. Artemisinin resistance and artemisinin-based combination therapy efficacy: status report. 2018. (2018). Pritchard, J. K., Pickrell, J. K. & Coop, G. The Genetics of Human Adaptation: Hard Sweeps, Soft Sweeps, and Polygenic Adaptation. Current Biology 20 , R208-R215 (2010). https://doi.org/10.1016/j.cub.2009.11.055 Amato, R. et al. Origins of the current outbreak of multidrug-resistant malaria in southeast Asia: a retrospective genetic study. Lancet Infect Dis 18 , 337-345 (2018). https://doi.org/10.1016/s1473-3099(18)30068-9 Miotto, O. et al. Genetic architecture of artemisinin-resistant Plasmodium falciparum. Nature Genetics 47 , 226-234 (2015). https://doi.org/10.1038/ng.3189 Michel, M. et al. Ancient Plasmodium genomes shed light on the history of human malaria. Nature (2024). https://doi.org/10.1038/s41586-024-07546-2 Molina-Cruz, A., Canepa, G. E. & Barillas-Mury, C. Plasmodium P47: a key gene for malaria transmission by mosquito vectors. Curr Opin Microbiol 40 , 168-174 (2017). https://doi.org/10.1016/j.mib.2017.11.029 Molina-Cruz, A. et al. Plasmodium falciparum evades immunity of anopheline mosquitoes by interacting with a Pfs47 midgut receptor. Proceedings of the National Academy of Sciences 117 , 2597-2605 (2020). https://doi.org/10.1073/pnas.1917042117 Marin-Mogollon, C. et al. The Plasmodium falciparum male gametocyte protein P230p, a paralog of P230, is vital for ookinete formation and mosquito transmission. Scientific Reports 8 , 14902 (2018). https://doi.org/10.1038/s41598-018-33236-x Auburn, S. et al. Characterization of Within-Host Plasmodium falciparum Diversity Using Next-Generation Sequence Data. PLOS ONE 7 , e32891 (2012). https://doi.org/10.1371/journal.pone.0032891 Manske, M. et al. Analysis of Plasmodium falciparum diversity in natural infections by deep sequencing. Nature 487 , 375-379 (2012). https://doi.org/10.1038/nature11174 Ngwana-Joseph, G. C. et al. Genomic analysis of global Plasmodium vivax populations reveals insights into the evolution of drug resistance. Nature Communications 15 , 10771 (2024). https://doi.org/10.1038/s41467-024-54964-x Guo, B. et al. Strong positive selection biases identity-by-descent-based inferences of recent demography and population structure in Plasmodium falciparum. Nature Communications 15 , 2499 (2024). https://doi.org/10.1038/s41467-024-46659-0 Birnbaum, J. et al. A Kelch13-defined endocytosis pathway mediates artemisinin resistance in malaria parasites. Science 367 , 51-59 (2020). https://doi.org/10.1126/science.aax4735 Winter, G. et al. SURFIN is a polymorphic antigen expressed on Plasmodium falciparum merozoites and infected erythrocytes. Journal of Experimental Medicine 201 , 1853-1863 (2005). https://doi.org/10.1084/jem.20041392 Trenholme, K. R. et al. clag9: A cytoadherence gene in Plasmodium falciparum essential for binding of parasitized erythrocytes to CD36. Proc Natl Acad Sci U S A 97 , 4029-4033 (2000). https://doi.org/10.1073/pnas.040561197 Sukkanon, C. et al. Prevalence of Plasmodium spp. in Anopheles mosquitoes in Thailand: a systematic review and meta-analysis. Parasit Vectors 15 , 285 (2022). https://doi.org/10.1186/s13071-022-05397-2 Sinka, M. E. et al. A global map of dominant malaria vectors. Parasit Vectors 5 , 69 (2012). https://doi.org/10.1186/1756-3305-5-69 Read, D. F., Cook, K., Lu, Y. Y., Le Roch, K. G. & Noble, W. S. Predicting gene expression in the human malaria parasite Plasmodium falciparum using histone modification, nucleosome positioning, and 3D localization features. PLOS Computational Biology 15 , e1007329 (2019). https://doi.org/10.1371/journal.pcbi.1007329 Malaria, G. E. N. P. f. C. P. Genomic epidemiology of artemisinin resistant malaria. eLife 5 , e08714 (2016). https://doi.org/10.7554/eLife.08714 Miotto, P. et al. A standardised method for interpreting the association between mutations and phenotypic drug resistance in Mycobacterium tuberculosis. Eur Respir J 50 (2017). https://doi.org/10.1183/13993003.01354-2017 Nayak, S. et al. Population genomics and transcriptomics of Plasmodium falciparum in Cambodia and Vietnam uncover key components of the artemisinin resistance genetic background. Nature Communications 15 , 10625 (2024). https://doi.org/10.1038/s41467-024-54915-6 Schmidt, S. et al. The Kelch13 compartment contains highly divergent vesicle trafficking proteins in malaria parasites. PLoS Pathog 19 , e1011814 (2023). https://doi.org/10.1371/journal.ppat.1011814 Ñacata, I., Early, A. M., Boboy, J., Neafsey, D. E. & Sáenz, F. E. Effects of drug pressure and human migration on antimalarial resistance in circulating Plasmodium falciparum malaria parasites in Ecuador. Res Sq (2024). https://doi.org/10.21203/rs.3.rs-4638168/v1 Coonahan, E. et al. Whole-genome surveillance identifies markers of Plasmodium falciparum drug resistance and novel genomic regions under selection in Mozambique. mBio 14 , e0176823 (2023). https://doi.org/10.1128/mbio.01768-23 Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30 , 2114-2120 (2014). https://doi.org/10.1093/bioinformatics/btu170 Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25 , 1754-1760 (2009). https://doi.org/10.1093/bioinformatics/btp324 Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25 , 2078-2079 (2009). https://doi.org/10.1093/bioinformatics/btp352 McKenna, A. et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res 20 , 1297-1303 (2010). https://doi.org/10.1101/gr.107524.110 Poplin, R. et al. Scaling accurate genetic variant discovery to tens of thousands of samples. bioRxiv , 201178 (2018). https://doi.org/10.1101/201178 Purcell, S. et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 81 , 559-575 (2007). https://doi.org/10.1086/519795 Alexander, D. H., Novembre, J. & Lange, K. Fast model-based estimation of ancestry in unrelated individuals. Genome Res 19 , 1655-1664 (2009). https://doi.org/10.1101/gr.094052.109 Schaffner, S. F., Taylor, A. R., Wong, W., Wirth, D. F. & Neafsey, D. E. hmmIBD: software to infer pairwise identity by descent between haploid genotypes. Malaria Journal 17 , 196 (2018). https://doi.org/10.1186/s12936-018-2349-7 Henden, L., Lee, S., Mueller, I., Barry, A. & Bahlo, M. Identity-by-descent analyses for measuring population dynamics and selection in recombining pathogens. PLOS Genetics 14 , e1007279 (2018). https://doi.org/10.1371/journal.pgen.1007279 Gautier, M. & Vitalis, R. rehh: an R package to detect footprints of selection in genome-wide SNP data from haplotype structure. Bioinformatics 28 , 1176-1177 (2012). https://doi.org/10.1093/bioinformatics/bts115 Mi, H. & Thomas, P. PANTHER pathway: an ontology-based pathway database coupled with data analysis tools. Methods Mol Biol 563 , 123-140 (2009). https://doi.org/10.1007/978-1-60761-175-2_7 Frichot, E. & François, O. LEA: An R package for landscape and ecological association studies. Methods in Ecology and Evolution 6 , 925-929 (2015). https://doi.org/https://doi.org/10.1111/2041-210X.12382 Szklarczyk, D. et al. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res 47 , D607-D613 (2019). https://doi.org/10.1093/nar/gky1131 Tables Table 1. Regional Summary of Plasmodium falciparum Genetic Diversity and Within-Host Complexity (N=17,565) Region N No. countries % F WS Median (Range) Median SNP Diversity (π) West Africa 5857 13 33.33 0.95 (0.23-1.00) 1.62 x 10 -3 Southeast Asia 5790 5 32.95 0.98 (0.43-1.00) 1.17 x 10 -3 East Africa 1604 6 9.13 0.95 (0.25-1.00) 1.66 x 10 -3 South America 1562 7 8.89 0.99 (0.60-1.00) 8.24 x 10 -4 South Asia 1554 2 8.84 0.97 (0.42-1.00) 1.51 x 10 -3 Central Africa 514 1 2.93 0.90 (0.31-1.00) 1.61 x 10 -3 Oceania 343 2 1.95 0.97 (0.48-1.00) 1.25 x 10 -3 South Central Africa 255 1 1.45 0.77 (0.32-1.00) 1.54 x 10 -3 Horn of Africa 92 2 0.52 0.99 (0.61-1.00) 1.51 x 10 -3 Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryFiguresPFPopulationGeneticsAnalysis.pdf Supplementary Figures SupplementaryTablesPFPopulationGeneticsAnalysis.xlsx Supplementary Tables Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7160640","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":494441484,"identity":"3ff8c623-e7c6-419d-834c-9a7e7c5595dc","order_by":0,"name":"Nina Billows","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBElEQVRIie3RMWrDMBSAYQmBsyjpquDQXsFBYBya3sXCQyfvnRxNmtQD9BaFQtcmvCFLIKvHHsEhEBSSoX5Olw5yMnbQP9gI+0N6NiGh0D8sgfZCnZiMcLXTlfqq88sTH8k6YgWP8P6mQZnJdz+ZL/FVQzrChnqpzMM1sh6uGmozHg2s3D8hiXPaOALSR1IYFaKdhUd8k8alrhYtYWNLIPUTnohuFlF+slIz3IXEhMDcRyRw6ajpyMd+hkTk7NRHpsBT8UveY3ohEe7iPRiSTOHB+PYwfsWPLJSZ2eTZO36y3ch656r7u4EpmuO5akkBtXt5nGqfwfK/S6p7/kooFAqFbukHlFhUononfVEAAAAASUVORK5CYII=","orcid":"","institution":"Faculty of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine","correspondingAuthor":true,"prefix":"","firstName":"Nina","middleName":"","lastName":"Billows","suffix":""},{"id":494441485,"identity":"eedf995d-8128-462a-a278-3bc5b8338e2f","order_by":1,"name":"Jamille G. Dombrowski","email":"","orcid":"","institution":"Department of Parasitology, Institute of Biomedical Sciences, University of São Paulo","correspondingAuthor":false,"prefix":"","firstName":"Jamille","middleName":"G.","lastName":"Dombrowski","suffix":""},{"id":494441486,"identity":"2fe67f98-88ec-4bdf-a3af-11358fe4d267","order_by":2,"name":"Joseph Thorpe","email":"","orcid":"","institution":"London School of Hygiene and Tropical Medicine","correspondingAuthor":false,"prefix":"","firstName":"Joseph","middleName":"","lastName":"Thorpe","suffix":""},{"id":494441487,"identity":"1f5ac2e8-18f6-4810-af5b-29ffb538c3a3","order_by":3,"name":"Leen Vanheer","email":"","orcid":"https://orcid.org/0000-0002-7680-385X","institution":"London School of Hygiene and Tropical Medicine","correspondingAuthor":false,"prefix":"","firstName":"Leen","middleName":"","lastName":"Vanheer","suffix":""},{"id":494441488,"identity":"f49aaa6d-c74d-40e7-8c4e-715412a55692","order_by":4,"name":"Sophie Moss","email":"","orcid":"","institution":"Faculty of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine","correspondingAuthor":false,"prefix":"","firstName":"Sophie","middleName":"","lastName":"Moss","suffix":""},{"id":494441489,"identity":"a4a09918-b459-42d0-96f3-c714b1465c38","order_by":5,"name":"Jesse Gitaka","email":"","orcid":"https://orcid.org/0000-0001-7516-522X","institution":"Directorate of Research and Innovation, Mount Kenya University","correspondingAuthor":false,"prefix":"","firstName":"Jesse","middleName":"","lastName":"Gitaka","suffix":""},{"id":494441490,"identity":"4294e865-c13b-43c3-b2b5-11f9ab6d9e9f","order_by":6,"name":"Colin Sutherland","email":"","orcid":"https://orcid.org/0000-0003-1592-6407","institution":"London School of Hygiene and Tropical Medicine","correspondingAuthor":false,"prefix":"","firstName":"Colin","middleName":"","lastName":"Sutherland","suffix":""},{"id":494441491,"identity":"a1ea0166-72bd-4ee3-96fd-ad7ef20558cf","order_by":7,"name":"Claudio R. F. Marinho","email":"","orcid":"","institution":"Department of Parasitology, Institute of Biomedical Sciences, University of São Paulo","correspondingAuthor":false,"prefix":"","firstName":"Claudio","middleName":"R. F.","lastName":"Marinho","suffix":""},{"id":494441492,"identity":"d80aeead-ca67-48a9-a5f3-bbb0110398ef","order_by":8,"name":"Nguyen Ngoc TH","email":"","orcid":"","institution":"Molecular Biology Department, Parasitology and Entomology, Vietnam National Institute of Malariology","correspondingAuthor":false,"prefix":"","firstName":"Nguyen","middleName":"Ngoc","lastName":"TH","suffix":""},{"id":494441493,"identity":"7af84a2a-7067-4403-9e12-5ca9488f71cf","order_by":9,"name":"Nguyen Binh TH","email":"","orcid":"","institution":"Molecular Biology Department, Parasitology and Entomology, Vietnam National Institute of Malariology","correspondingAuthor":false,"prefix":"","firstName":"Nguyen","middleName":"Binh","lastName":"TH","suffix":""},{"id":494441494,"identity":"21d8dc82-65d2-4b02-b4c3-3150f526123b","order_by":10,"name":"Nguyen Thieu Q","email":"","orcid":"","institution":"Molecular Biology Department, Parasitology and Entomology, Vietnam National Institute of Malariology","correspondingAuthor":false,"prefix":"","firstName":"Nguyen","middleName":"Thieu","lastName":"Q","suffix":""},{"id":494441495,"identity":"b3c6b4c3-30fa-4c20-9c90-94fe9b52a9c1","order_by":11,"name":"Susana Campino","email":"","orcid":"https://orcid.org/0000-0003-1403-6138","institution":"London School of Hygiene and Tropical Medicine","correspondingAuthor":false,"prefix":"","firstName":"Susana","middleName":"","lastName":"Campino","suffix":""},{"id":494441496,"identity":"7b6d7106-1ca9-4675-ac50-f47ea51c71c8","order_by":12,"name":"Taane Clark","email":"","orcid":"https://orcid.org/0000-0001-8985-9265","institution":"London School of Hygiene \u0026 Tropical Medicine","correspondingAuthor":false,"prefix":"","firstName":"Taane","middleName":"","lastName":"Clark","suffix":""}],"badges":[],"createdAt":"2025-07-18 21:10:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7160640/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7160640/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89269092,"identity":"6b6379f7-7738-4168-94d9-0e5b6b246a8e","added_by":"auto","created_at":"2025-08-18 08:35:54","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":603735,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGlobal Population structure and admixture of P. falciparum (n=17,565).\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMultidimensional Scaling (MDS): Dimension 1 (D1) vs. Dimension 2 (D2) (A) demonstrate clustering across geographical regions. PCs are coloured according to each sample’s respective region. Admixture: Admixture analysis of samples from West Africa, Horn of Africa, Central Africa, East Africa, South Central Africa, South America, Southeast Asia, South Asia and Oceania. MDS of \u003cem\u003eP. falciparum\u003c/em\u003e coloured according to the maximum K value of each isolate (B). Ancestry coefficients (y-axis) are plotted for each isolate (x-axis) belonging to each region for K=10 ancestral populations (C).\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-7160640/v1/50f7217410b63b99a8665826.png"},{"id":89269093,"identity":"e394f7d7-6725-4b0c-aa8d-432ce9ac8312","added_by":"auto","created_at":"2025-08-18 08:35:54","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":544077,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGenetic differentiation (F\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003eST\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e) is greater between intercontinental regions. \u003c/strong\u003eA network map showing edges connecting regions weighted by average difference in F\u003csub\u003eST\u003c/sub\u003e scores between regions.\u0026nbsp;\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-7160640/v1/bf9dd8f4f221de229c9a7673.png"},{"id":89270229,"identity":"23322f97-5cf3-4e07-b7ce-c4af5e225723","added_by":"auto","created_at":"2025-08-18 08:43:54","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":403578,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProportion and distribution of drug-resistant genotypes across the global \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eP. falciparum \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003edataset (N=17,565). \u003c/strong\u003eDrug-resistant genotypes were defined according to the malaria-profiler and WHO databases. Pie charts represent the proportion of drug-resistant genotype patterns represented by a colour across each country. None indicates either drug-susceptibility or missing genotypes due to lack of genome coverage.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-7160640/v1/3286a84d1e496fc0b48ffe0f.png"},{"id":89269095,"identity":"919a9759-e7bc-4d31-b493-5841a8969bb0","added_by":"auto","created_at":"2025-08-18 08:35:54","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":353040,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIBD networks illustrating clonal expansion of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003epfkelch13\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e mutant lineages from 2008 to 2019 across Southeast Asia. \u003c/strong\u003eIBD networks were constructed using a 95% pairwise IBD threshold for three time periods: 2008–2011 (A), 2012–2015 (B), and 2016–2019 (C). Nodes represent parasite samples and are coloured by \u003cem\u003epfkelch13 \u003c/em\u003egenotype: C580Y, other missense mutations in the propeller domain, or wild-type. Samples with missing \u003cem\u003epfkelch13\u003c/em\u003e genotypes were excluded. Boxplot showing the distribution of pairwise IBD fractions among samples grouped by \u003cem\u003epfkelch13 \u003c/em\u003egenotype: C580Y, other propeller domain mutations, and wild-type (D). Pie chart showing the proportion of \u003cem\u003epfkelch13 \u003c/em\u003egenotypes (C580Y, other mutations, wild-type) by country (Laos, Cambodia, Vietnam, Myanmar, Thailand). The colour key from this panel is used consistently across the figure (E). Protein–protein interaction network of genes containing SNPs significantly associated with \u003cem\u003epfkelch13\u003c/em\u003e C580Y. Nodes are grouped by k-means clustering and coloured by cluster; edge width indicates interaction confidence (ranging from 0.4 to 1.0) (F).\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-7160640/v1/68efe7d1561c3a0ac7699531.png"},{"id":89270576,"identity":"8182c47c-d483-4a12-9e6d-07d4ccf07a32","added_by":"auto","created_at":"2025-08-18 08:51:56","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3673536,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7160640/v1/c6208804-a6dd-4efd-b862-1a99888fd809.pdf"},{"id":89269094,"identity":"12dd07d6-547f-41e8-9449-2a4040b9935a","added_by":"auto","created_at":"2025-08-18 08:35:54","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1963629,"visible":true,"origin":"","legend":"Supplementary Figures","description":"","filename":"SupplementaryFiguresPFPopulationGeneticsAnalysis.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7160640/v1/0d766edd9b78ec405ce5f6ee.pdf"},{"id":89269097,"identity":"1b86dad3-d66b-4482-bde4-2967750ab643","added_by":"auto","created_at":"2025-08-18 08:35:54","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":4070947,"visible":true,"origin":"","legend":"Supplementary Tables","description":"","filename":"SupplementaryTablesPFPopulationGeneticsAnalysis.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7160640/v1/5c44b67883c16a5024216b08.xlsx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Global-scale population genetic analysis of Plasmodium falciparum identifies country- and region-specific patterns of malaria parasite adaptation","fulltext":[{"header":"BACKGROUND","content":"\u003cp\u003eMalaria, a vector-borne disease caused by \u003cem\u003ePlasmodium\u003c/em\u003e parasites and transmitted by \u003cem\u003eAnopheles\u003c/em\u003e mosquitoes, remains one of the most significant global public health challenges. Among the six \u003cem\u003ePlasmodium\u003c/em\u003e species that infect humans, \u003cem\u003ePlasmodium falciparum\u003c/em\u003e causes the most severe disease and accounts for the majority of malaria-related deaths, particularly among young children in sub-Saharan Africa. In 2023 alone, global estimates reached 263 million malaria cases and close to 600,000 deaths, underscoring the persistent burden of this disease despite decades of control efforts\u0026nbsp;\u003csup\u003e1\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eWhile progress has been made through antimalarial treatments and vector control strategies, multiple biological threats have emerged that compromise their effectiveness \u003csup\u003e2,3\u003c/sup\u003e. These include the evolution of drug resistance in the parasite and insecticide resistance in mosquito vectors. In addition, the influence of global warming is increasingly recognised as a major factor shaping malaria transmission patterns, altering seasonality, changing human migration patterns and so potentially expanding the geographic range of the disease \u003csup\u003e4\u003c/sup\u003e. Together, these challenges underscore the need for a coordinated global response to control malaria and progress towards its eradication. A critical component of this response is addressing the biological processes that undermine current interventions. In particular, understanding the genetic diversity and evolutionary dynamics of \u003cem\u003eP. falciparum\u003c/em\u003e enables tracking of the emergence and spread of drug resistance. While much of this work has focused on national or regional scales, expansion of efforts to map the evolutionary trajectories of \u003cem\u003eP. falciparum\u003c/em\u003e to a global level can anticipate resistance trends and hasten an effective response to waning effectiveness of malaria drug regimens.\u003c/p\u003e\n\u003cp\u003eThe evolution of \u003cem\u003eP. falciparum\u003c/em\u003e reflects a complex interplay of ecological, biological and anthropogenic factors. In addition, malaria control interventions have imposed strong selective pressures, causing population bottlenecks and regional differences in diversity \u003csup\u003e5-9\u003c/sup\u003e. Regional differences in \u003cem\u003eP. falciparum\u003c/em\u003e population structure have been well documented. African populations show high diversity due to intense transmission, while lower diversity is seen in South America and intermediate levels in Southeast Asia. These patterns align with region-specific antimalarial drug use and resistance histories \u003csup\u003e10-13\u003c/sup\u003e. Key resistance markers, such as mutations in \u003cem\u003epfcrt\u003c/em\u003e (chloroquine resistance transporter; chloroquine), \u003cem\u003epfmdr1\u003c/em\u003e (multi-drug resistance; amodiaquine, chloroquine, mefloquine, piperaquine), \u003cem\u003epfdhfr\u003c/em\u003e (dihydrofolate reductase; pyrimethamine), d\u003cem\u003efdhps\u003c/em\u003e (dihydropteroate synthase; sulfadoxine) and \u003cem\u003epfkelch13\u003c/em\u003e (Kelch propeller protein; artemisinin), exhibit geographic variation, but current panels remain incomplete \u003csup\u003e14-18\u003c/sup\u003e. These resistance markers include both single-nucleotide polymorphisms (SNPs) and structural or copy number variants, all of which exhibit regional variation in prevalence and pattern \u003csup\u003e19\u003c/sup\u003e. Continued identification of novel mutations is critical for surveillance and guiding treatment \u003csup\u003e20\u003c/sup\u003e. Artemisinin resistance, first emerging in Southeast Asia, now poses a growing threat in Africa, with concerning signs of reduced ACT partner drug efficacy, such as lumefantrine resistance in Uganda \u003csup\u003e21,22\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003ePopulation genomics is a powerful tool to dissect the genetic landscape of \u003cem\u003eP. falciparum\u003c/em\u003e and uncover the mechanisms driving its evolution and adaptation. By combining classical statistical techniques and high-resolution genomic analyses, researchers can quantify nucleotide diversity, identify loci associated with drug resistance and population divergence, characterise population structure and genetic relatedness, and unravel complex infection patterns \u003csup\u003e23-26\u003c/sup\u003e. These approaches have significantly advanced our understanding of the connectivity of parasite populations, provided insights into gene flow and admixture, and shown how quickly \u003cem\u003eP. falciparum\u003c/em\u003e can adapt in response to selective pressures \u003csup\u003e27-29\u003c/sup\u003e. Importantly, population genetics has provided a framework for inferring the origins and spread of drug-resistant mutations, as well as for analysing diversity in genes encoding vaccine targets, both of which are major contributors to intervention response \u003csup\u003e30\u003c/sup\u003e. The expansion of publicly accessible genomic datasets has further enhanced the scale and resolution of such analyses. Notably, the 2023 release of the Pf7 data resource by MalariaGEN included nearly 20,000 \u003cem\u003eP. falciparum\u0026nbsp;\u003c/em\u003ewhole genome sequences from 33 countries \u003csup\u003e26\u003c/sup\u003e. A number of research groups have also made substantial contributions to the growing repository of genomic data, including from formerly underrepresented malaria-endemic countries, further enabling large-scale comparative analyses and global surveillance efforts \u003csup\u003e31-37\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eLeveraging the growing wealth of publicly accessible genomic data, alongside newly generated whole-genome sequencing (WGS) data from Brazil and Vietnam, this study employs large-scale SNP-based analyses to approach \u003cem\u003eP. falciparum\u0026nbsp;\u003c/em\u003epopulation dynamics from a global perspective. Specifically, the contribution of genetic variation to regional differences in infection complexity, parasite diversity and the emergence and dissemination of antimalarial drug resistance is investigated. By comparing transmission dynamics and evolutionary pressures across diverse geographic settings, the response of \u003cem\u003eP. falciparum\u003c/em\u003e to local intervention strategies can be examined. Importantly, the approach also explores the genomic architecture of resistance, including evidence for polygenic interactions that may drive adaptive responses. These findings reveal how regional drug use and transmission dynamics shape\u003cem\u003e\u0026nbsp;P. falciparum\u003c/em\u003e evolution, underscoring the need for region-specific genetic surveillance and treatment policies to effectively manage resistance and support malaria elimination.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003e\u003cstrong\u003eGenomic and Geographic Diversity of \u003cem\u003eP. falciparum\u0026nbsp;\u003c/em\u003eisolates\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAfter quality control procedures, the final dataset comprised 17,565 \u003cem\u003eP. falciparum\u003c/em\u003e isolates from 39 countries across 9 regions (Africa (Central, East, Horn, West, South Central), Asia (South, Southeast), South America, Oceania; \u003cstrong\u003eTable 1\u003c/strong\u003e). Most isolates originated from West Africa (33.3%) and Southeast Asia (33.0%), with smaller percentages from East Africa (9.1%), South America (8.9%), South Asia (8.8%), Central Africa (2.9%), Oceania (2.0%), South Central Africa (1.5%), and the Horn of Africa (0.5%). Notably, Ghana contributed a significant portion of the isolates (n=2,845, 16.2%), while each of the other countries contributed less than 10% to the total sample set (\u003cstrong\u003eSupplementary Table 1\u003c/strong\u003e). High-quality genome-wide variants were identified (3,600,488 SNPs in total) and 1,717,351 (47.7%) biallelic SNPs (population-specific minor allele frequency (MAF)\u0026gt;0.001) met the criteria for further analysis. Most shortlisted SNPs were non-synonymous mutations (57.5%). Overall, SNP diversity was comparable across all regions (median SNP \u0026pi; range: 8.24 \u0026times; 10⁻⁴ \u0026ndash; 1.66 \u0026times; 10⁻\u0026sup3;), with slightly lower estimates for South American samples (\u003cstrong\u003eTable 1\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePopulation structure and Admixture\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMultidimensional scaling (MDS) revealed that isolates clustered according to their respective geographical regions and countries (\u003cstrong\u003eFig. 1\u003c/strong\u003e). Significant overlap was observed among the African regions (Central Africa, West Africa, East Africa, Horn of Africa and South Central Africa) as well as among the Asian/Oceanic regions (South Asia, Southeast Asia and Oceania) (\u003cstrong\u003eFig. 1\u003c/strong\u003e). South American \u003cem\u003eP. falciparum\u003c/em\u003e isolates display distinct regional structuring, reflecting geographic separation and historical transmission (\u003cstrong\u003eSupplementary Figure 1\u003c/strong\u003e). Colombian and Ecuadorian isolates form a distinct cluster that is relatively close to African samples along the primary axis of genetic variation. However, they show increased separation along secondary components that capture additional population structure, highlighting subtle divergence. Brazilian isolates are more dispersed, overlapping with African, Colombian and other South American clusters. In contrast, isolates from Peru, French Guiana, Guyana and Venezuela group closely together to form a separate South American cluster. All South American and African clusters are clearly distinct from those in Southeast Asia, South Asia and Oceania (\u003cstrong\u003eSupplementary Figure 1\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eTo further investigate the geographical origins of \u003cem\u003eP. falciparum\u003c/em\u003e isolates, ancestry inference was conducted using maximum likelihood estimation via ADMIXTURE analysis, based on 974,060 SNPs with a MAF greater than 1%. The optimal number of ancestral populations (K = 10) was determined using cross-validation. The ADMIXTURE results revealed patterns of shared ancestry consistent with known geographic origins. For instance, four dominant ancestral clusters were observed among African isolates, with West Africa containing a distinct ancestral component compared to other African regions (\u003cstrong\u003eFig. 1\u003c/strong\u003e). Similarly, Southeast Asian isolates exhibited four dominant ancestral populations, with shared components also observed with isolates from Oceania, reflected in three overlapping ancestral clusters. Several Southeast Asian samples showed signatures of East African ancestry, supporting hypotheses of both the African origin of \u003cem\u003eP. falciparum\u003c/em\u003e and human migration from Africa to Southeast Asia \u003csup\u003e38\u003c/sup\u003e. Isolates from South America and Southeast Asia exhibited more distinct and region-specific ancestry profiles relative to other parts of the world. At the country level, ancestral compositions were generally consistent within national borders, with overlapping patterns observed in neighbouring countries (\u003cstrong\u003eSupplementary Figures 2\u0026ndash;3\u003c/strong\u003e). In contrast, isolates from Colombia and Ecuador had a unique ancestral composition, aligning with MDS findings and suggesting multiple introductions of \u003cem\u003eP. falciparum\u003c/em\u003e into South America during historical periods of colonisation \u003csup\u003e5\u003c/sup\u003e. The Horn of Africa has also exhibited some limited gene flow from South Asia, as well as mainly Central-East African gene flow (\u003cstrong\u003eSupplementary Figure 4\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMultiplicity of Infection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe F\u003csub\u003eWS\u003c/sub\u003e statistic was employed to summarise relative inbreeding and multiplicity of infection (MOI), where higher values (closer to 1) indicate clonality, and lower values (closer to 0) reflect a mixture of unrelated clones. An F\u003csub\u003eWS\u0026nbsp;\u003c/sub\u003evalue greater than 0.95 typically signifies a monoclonal population. Regionally, the highest median F\u003csub\u003eWS\u003c/sub\u003e values were observed in South America (0.99), Horn of Africa (0.99), Southeast Asia (0.98) and Oceania (0.97) (\u003cstrong\u003eTable 1,\u003c/strong\u003e \u003cstrong\u003eSupplementary Figure 5\u003c/strong\u003e). These regions also had a higher percentage of samples with F\u003csub\u003eWS\u003c/sub\u003e \u0026gt; 0.95, indicating low MOI. In contrast, lower mean F\u003csub\u003eWS\u0026nbsp;\u003c/sub\u003evalues and a smaller percentage of samples with F\u003csub\u003eWS\u003c/sub\u003e \u0026gt; 0.95 were reported for South Central Africa (0.74), Central Africa (0.82), West Africa (0.85), East Africa (0.85), and South Asia (0.90) (\u003cstrong\u003eSupplementary Figure 5\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAt the country level, F\u003csub\u003eWS\u0026nbsp;\u003c/sub\u003evalues fluctuated over time which could suggest changes in transmission intensity. Statistically significant changes (P \u0026lt; 0.05) in MOI across year groups were observed in Bangladesh (ANOVA, P = 0.025), Cambodia (P = 0.025), Colombia (P = 0.021), The Gambia (P \u0026lt; 2\u0026times;10⁻\u0026sup1;⁶), Ghana (P = 6.07\u0026times;10⁻⁶), Kenya (P = 1.98\u0026times;10⁻⁸), Laos (P = 2.31\u0026times;10⁻⁹), Mali (P = 1\u0026times;10⁻⁴), Myanmar (P = 0.001), Papua New Guinea (P = 0.003), Thailand (P = 9.21\u0026times;10⁻⁵) and Vietnam (P = 2.33\u0026times;10⁻⁹) (\u003cstrong\u003eSupplementary Table 2\u003c/strong\u003e). In several countries, rising F\u003csub\u003eWS\u003c/sub\u003e scores over time may indicate declining transmission or increased sampling in areas of lower endemicity. Notable examples include Bangladesh, with a median F\u003csub\u003eWS\u003c/sub\u003e increasing from 0.89 (range: 0.45\u0026ndash;1.00) in 2008\u0026ndash;2011 to 0.97 (range: 0.45\u0026ndash;1.00) in 2016\u0026ndash;2019; The Gambia, from 0.74 (range: 0.25\u0026ndash;1.00) pre-2000 to 0.97 (range: 0.43\u0026ndash;1.00) in 2016\u0026ndash;2019; Ghana, from 0.86 (range: 0.29\u0026ndash;1.00) 2008\u0026ndash;2011 to 0.89 (range: 0.25\u0026ndash;1.00) 2016\u0026ndash;2019; and Laos, from 0.97 in 2008\u0026ndash;2011 (range: 0.54\u0026ndash;1.00) to 0.99 in 2016\u0026ndash;2019 (range: 0.49\u0026ndash;1.00). In other countries, F\u003csub\u003eWS\u003c/sub\u003e scores showed variability and modest declines over time, potentially indicating rising transmission intensity. This pattern was observed in Benin, Cambodia, Guyana, Indonesia, Papua New Guinea and Vietnam (\u003cstrong\u003eSupplementary Table 2\u003c/strong\u003e). Notably, some countries experienced more pronounced declines in F\u003csub\u003eWS\u0026nbsp;\u003c/sub\u003escores in recent years compared to earlier periods. For example, Mali saw a drop from a median F\u003csub\u003eWS\u0026nbsp;\u003c/sub\u003eof 0.93 (range: 0.34\u0026ndash;1.00) in 2016\u0026ndash;2019 to 0.73 (range: 0.48\u0026ndash;0.99) in 2020\u0026ndash;2021. Similarly, Kenya\u0026rsquo;s scores decreased from a median of 0.98 (range: 0.26\u0026ndash;1.00) in 2012\u0026ndash;2015 to 0.87 (range: 0.48\u0026ndash;1.00) in 2016\u0026ndash;2019. These shifts could reflect substantial increases in transmission or a focus on sampling in higher-transmission areas.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGenomic Relatedness\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIdentity by descent (IBD) analysis was performed to identify genomic segments shared between \u003cem\u003eP. falciparum\u0026nbsp;\u003c/em\u003eisolates that have not undergone recombination, serving as markers of inheritance from a recent common ancestor and providing a measure of genomic relatedness. The fraction of the genome classified with IBD was calculated using 10 kb sliding windows and median IBD fractions (with ranges) were initially summarised by regional groupings (\u003cstrong\u003eSupplementary Figure 6\u003c/strong\u003e). The highest IBD fractions were observed in samples from South America (median: 0.150; range: 0.003\u0026ndash;0.663), followed by Southeast Asia (median: 0.055; range: 0.001\u0026ndash;0.238), the Horn of Africa (median: 0.029; range: 0.001\u0026ndash;0.227), Oceania (median: 0.028; range: 0.001\u0026ndash;0.212), and South Central Africa (median: 0.012; range: 0.001\u0026ndash;0.141). In contrast, the lowest IBD fractions were seen in South Asia (median: 0.002; range: \u0026lt;0.001\u0026ndash;0.119), East Africa (median: 0.001; range: \u0026lt;0.001\u0026ndash;0.114), Central Africa (median: 0.001; range: \u0026lt;0.001\u0026ndash;0.114), and West Africa (median: 0.001; range: \u0026lt;0.001\u0026ndash;0.100). These findings are consistent with patterns observed in F\u003csub\u003eWS\u003c/sub\u003e scores and reflect the impact of higher transmission intensity in these regions. Similar trends were observed for pairwise IBD fractions, with South American samples exhibiting the highest levels of pairwise genomic relatedness (\u003cstrong\u003eSupplementary Figure 7\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eRegions with higher IBD fractions may indicate reduced outcrossing within parasite populations, typically associated with low transmission intensity or geographic isolation. Conversely, lower IBD fractions suggest increased outcrossing due to higher transmission and greater genetic mixing between isolates. To further explore this, genomic regions with the highest IBD fractions were examined to identify associated genes and gene products (\u003cstrong\u003eSupplementary Table 3\u003c/strong\u003e). A total of 377 high-IBD segments (top 5% of IBD values) were identified, encompassing 173 genes. The greatest number of high-IBD segments was observed in the Horn of Africa and Oceania (n = 125 each), followed closely by East Africa (n = 122), South America (n = 121), South Asia and Southeast Asia (n = 120 each), West Africa (n = 120), Central Africa (n = 119) and South Central Africa (n = 117). Several segments were conserved across multiple regions (\u003cstrong\u003eSupplementary Figure 8\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eGene Ontology (GO) term overrepresentation analysis showed that high-IBD regions across all geographic areas were significantly enriched for genes involved in protein\u0026ndash;DNA complex subunit organisation, chromatin and chromosome organisation, and the regulation of transcription, translation, gene expression and biosynthetic processes (\u003cstrong\u003eSupplementary Table 4\u003c/strong\u003e). These genes were primarily associated with chromosomal regions (fold change: 8.25; n = 5) and the nucleus (fold change: 1.59; n = 49) and exhibited molecular functions largely related to binding and structural molecule activity. Of particular interest, several high-IBD regions overlapped with loci associated with antimalarial drug resistance. These included \u003cem\u003epfcrt\u003c/em\u003e (in Central Africa, East Africa, Horn of Africa, Oceania, South America, South Asia, and West Africa), \u003cem\u003epfdhps\u003c/em\u003e (in East Africa, Horn of Africa, Oceania, South America, South Asia, South Central Africa, and Southeast Asia), and \u003cem\u003epfmdr1\u003c/em\u003e (in Central Africa, East Africa, Oceania, and South America), potentially reflecting drug-driven selective pressures. \u0026nbsp;Elevated IBD was also found in regions encoding Kelch13-interacting candidate genes \u003cem\u003eKIC7\u003c/em\u003e (Horn of Africa and South America) and \u003cem\u003eKIC9\u003c/em\u003e (South America), suggesting possible signatures of positive selection. Additionally, genes essential for parasite transmission and interaction with \u003cem\u003eAnopheles\u003c/em\u003e mosquitoes, \u003cem\u003eP47\u003c/em\u003e (involved in immune evasion) and \u003cem\u003eP48/45\u003c/em\u003e (linked to gamete fertility), showed high IBD in Oceania (\u003cstrong\u003eSupplementary Table 3\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePopulation Differentiation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo further investigate the genetic differentiation of \u003cem\u003eP. falciparum\u003c/em\u003e across global populations, fixation index (F\u003csub\u003eST\u003c/sub\u003e) analysis was conducted to identify highly differentiated genomic sites and assess patterns of divergence between regional subpopulations. A strong positive correlation was observed between genetic differentiation and geographic distance (Mantel r = 0.89, P = 0.001) (\u003cstrong\u003eFig. 2\u003c/strong\u003e). The lowest levels of genetic differentiation were observed within Africa (genome-wide F\u003csub\u003eST\u003c/sub\u003e range: 0.009\u0026ndash;0.071), as well as between South and Southeast Asia (genome-wide F\u003csub\u003eST\u003c/sub\u003e: 0.083). In contrast, the greatest differentiation was observed between Southeast Asia and South America (genome-wide F\u003csub\u003eST\u003c/sub\u003e: 0.365), reflecting the large geographic separation between these regions (~18,327 km) and historically low human population movement between them (\u003cstrong\u003eFig. 2\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003ePatterns of moderate to high differentiation (F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.75 and F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.95) also aligned with geographic distance. The highest number of highly differentiated SNPs (F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.75) were observed in comparisons involving South America: with Southeast Asia (n = 172, including 41 with F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.95), South Asia (n = 159, including 10 with F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.95) and Oceania (n = 82, including 32 with F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.95), supporting the influence of geographic distance on genotypic divergence. In contrast, no SNPs with extreme differentiation (F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.95) were identified between South and Southeast Asia, consistent with their relative proximity (average sampling distance: 1,722 km) and long-established human migration. Within Africa, only 14 highly differentiated sites (F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.75) were identified across five genes: \u003cem\u003eRhopH2\u003c/em\u003e, \u003cem\u003ePF3D7_1361800\u003c/em\u003e, \u003cem\u003ePF3D7_0811600\u003c/em\u003e, \u003cem\u003eGcalpha\u003c/em\u003e, and \u003cem\u003ePF3D7_0526600\u0026nbsp;\u003c/em\u003e(\u003cstrong\u003eSupplementary Table 5\u003c/strong\u003e), reflecting the lower overall genetic structure across the continent.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA total of 1,346 SNPs exhibited high genetic differentiation (F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.75) in intercontinental comparisons (\u003cstrong\u003eSupplementary Table 5\u003c/strong\u003e), highlighting key genomic regions underlying geographical divergence, including loci implicated in drug resistance and parasite transmission. Noteworthy examples include \u003cem\u003epfdhfr\u003c/em\u003e (pyrimethamine resistance), \u003cem\u003epfmdr1\u003c/em\u003e (chloroquine and mefloquine resistance), and \u003cem\u003epfcrt\u003c/em\u003e (chloroquine resistance), all of which harbour mutations known to mediate resistance phenotypes. Although no significant GO terms were identified for biological processes or molecular functions, sites with F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.75 were found within genes localised to cellular components, including the intrinsic component of the external side of the plasma membrane, anchored components of the plasma membrane, the apical complex, cell surface and apical part of the cell (\u003cstrong\u003eSupplementary Table 6\u003c/strong\u003e). Genes from multiple families also contained highly differentiated sites, particularly those involved in lipid scavenging, such as the ACS family (n = 38) and \u003cem\u003eACS10\u003c/em\u003e (n = 34), which have been proposed as drug targets, but are also highly variable \u003csup\u003e39\u003c/sup\u003e. Additional high-F\u003csub\u003eST\u003c/sub\u003e sites were observed in genes mediating parasite-host and parasite-vector interactions, including \u003cem\u003eP47\u0026nbsp;\u003c/em\u003e(n = 33), \u003cem\u003eCTRP\u003c/em\u003e (n = 20), and \u003cem\u003ePfs16\u003c/em\u003e (n = 9), or genes implicated in immune evasion, cellular invasion, or gametocyte development, and are considered vaccine candidates \u003csup\u003e26,40\u003c/sup\u003e. Comparable patterns were observed in country-level pairwise F\u003csub\u003eST\u003c/sub\u003e analyses (\u003cstrong\u003eSupplementary Table 7\u003c/strong\u003e). Interestingly, despite their geographical proximity, Brazil and Ecuador showed a relatively high number of highly differentiated sites (F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.75), a pattern more likely attributable to small sample sizes rather than true biological divergence. However, it could also reflect both geographic and sociolinguistic barriers to gene flow, including the lack of a shared border, distinct coastlines (Atlantic vs. Pacific) and differing colonial and linguistic histories that may limit parasite movement via human migration (\u003cstrong\u003eSupplementary Figure 1\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eFurther insights into regional differentiation were obtained using a \u0026apos;one-against-all\u0026apos; F\u003csub\u003eST\u003c/sub\u003e approach. At the regional level, only 30 sites showed high genetic differentiation (F\u003csub\u003eST\u0026nbsp;\u003c/sub\u003e\u0026gt; 0.75) and none exceeded an F\u003csub\u003eST\u0026nbsp;\u003c/sub\u003eof 0.95 (\u003cstrong\u003eSupplementary Table 8\u003c/strong\u003e). Most of these high F\u003csub\u003eST\u003c/sub\u003e sites were specific to South American isolates (n = 17), with fewer identified in West Africa (n = 9), Oceania (n = 3) and Southeast Asia (n = 1). These loci were located in genes associated with drug resistance (\u003cem\u003epfdhfr\u003c/em\u003e, Oceania), regulation of biological processes (\u003cem\u003eCRK3\u003c/em\u003e, \u003cem\u003eSET1\u003c/em\u003e), cellular metabolism (\u003cem\u003eACS10\u003c/em\u003e, \u003cem\u003ePF3D7_0709700\u003c/em\u003e), biosynthesis (\u003cem\u003ePF3D7_0713600\u003c/em\u003e, \u003cem\u003ePAIP1\u003c/em\u003e), reproduction (\u003cem\u003ePF3D7_0809600\u003c/em\u003e, \u003cem\u003eGIG\u003c/em\u003e), nutrient uptake (\u003cem\u003eRhopH2\u003c/em\u003e), cell localisation (\u003cem\u003ePF3D7_1440800\u003c/em\u003e, \u003cem\u003eHSP101\u003c/em\u003e), signalling (\u003cem\u003eGCbeta\u003c/em\u003e), cell binding (\u003cem\u003ePF3D7_1410400\u003c/em\u003e), and interspecies interactions (\u003cem\u003eP47\u003c/em\u003e), as well as other genes with diverse functions (\u003cem\u003ePF3D7_1116800\u003c/em\u003e, \u003cem\u003ePF3D7_1135100\u003c/em\u003e, \u003cem\u003ePF3D7_1442200\u003c/em\u003e). At the country level, the \u0026apos;one-against-all\u0026apos; analysis identified 211 high F\u003csub\u003eST\u003c/sub\u003e sites, comprising 132 unique SNPs with F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.75 and two sites with F\u003csub\u003eST\u003c/sub\u003e \u0026gt; 0.95 (\u003cstrong\u003eSupplementary Table 9\u003c/strong\u003e). Most of these sites were observed in isolates from Brazil, Ecuador, Colombia, Peru, and Indonesia, underscoring localised genetic structure. Several of these high F\u003csub\u003eST\u003c/sub\u003e sites overlapped with previously described genes and regions known to be highly variable, such as the \u003cem\u003eSURFIN\u003c/em\u003e family, \u003cem\u003epfcrt\u003c/em\u003e, and \u003cem\u003epfdhfr\u003c/em\u003e. Notably, isolates from Brazil, French Guiana, Peru, and Indonesia also carried highly differentiated sites in \u003cem\u003eFIKK4.2\u003c/em\u003e and \u003cem\u003eFIKK10.1\u003c/em\u003e, genes previously implicated in parasite invasion and virulence \u003csup\u003e41\u003c/sup\u003e. Together, these results emphasise region- and country-specific patterns of genetic differentiation in \u003cem\u003eP. falciparum\u003c/em\u003e, likely driven by a combination of geographic isolation, local selective pressures, and differing transmission dynamics.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGenomic regions under recent positive selection\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo investigate recent positive selection across the \u003cem\u003eP. falciparum\u003c/em\u003e genome, we applied within-population (iHS) and between-population (XP-EHH) analyses based on extended haplotype homozygosity (EHH). For iHS, SNPs exceeding the threshold of (\u0026minus; log10[1 \u0026ndash; 2 | \u0026Phi;iHS \u0026ndash; 0.5 |])\u0026thinsp;\u0026gt;\u0026thinsp;4.0 were under recent positive selection within single regions (\u003cstrong\u003eSupplementary Figure 9; Supplementary Table 10\u003c/strong\u003e). In total, 561 unique SNPs spanning 170 genes met this criterion. The highest number of sites were observed in West Africa (n = 207), followed by East Africa (n = 190), South Asia (n = 169) and Central Africa (n = 144), while fewer than 30 SNPs were detected in other regions. Most of the positively selected sites were found in genes encoding surface antigens or located in highly variable genomic regions, consistent with immune-mediated selection and adaptation for efficient merozoite invasion. These included \u003cem\u003eDBLMSP2\u003c/em\u003e, \u003cem\u003eTRAP\u003c/em\u003e, \u003cem\u003eCLAG8\u003c/em\u003e, \u003cem\u003eAMA1\u003c/em\u003e, \u003cem\u003eMSP1\u003c/em\u003e, and members of the \u003cem\u003eSURFIN\u003c/em\u003e family. Notably, these genes were enriched in the KEGG pathway pfa05144 (adjusted P = 0.001), which is associated with sporozoite invasion. In addition, 16 SNPs under positive selection were detected in genes involved in lipid metabolism, particularly \u003cem\u003eACS\u003c/em\u003e and \u003cem\u003eACS7\u003c/em\u003e, across East, West, Central Africa and Oceania. These acetyl-CoA synthetase genes play a key role in scavenging host fatty acids to support parasite growth \u003csup\u003e39\u003c/sup\u003e. Drug-resistance loci also showed signs of selection. For example, a non-synonymous mutation in \u003cem\u003epfcrt\u003c/em\u003e (405600C\u0026gt;T, Ile356Thr) was identified in West African populations, potentially reflecting historical chloroquine pressure or adaptation to current combination therapies. To identify regional selection hotspots, we further prioritised genes containing \u0026gt;5 SNPs exceeding the selection threshold, as well as the top five strongest selection signals per region (\u003cstrong\u003eSupplementary\u003c/strong\u003e \u003cstrong\u003eFigure 9\u003c/strong\u003e). In African populations, notable hotspots included \u003cem\u003ePF3D7_0809600\u003c/em\u003e (C50 cysteine protease) and \u003cem\u003ePF3D7_1475800\u003c/em\u003e (hypothetical protein) in East Africa; \u003cem\u003ePF3D7_1028000\u003c/em\u003e (uncharacterised, also selected in West Africa) in the Horn of Africa; and \u003cem\u003ePF3D7_0113800\u0026nbsp;\u003c/em\u003e(DBL-containing protein) in South Central Africa. Outside of Africa, strong selection signals were detected in \u003cem\u003ePF3D7_1475900\u003c/em\u003e (KELT protein, South Asia), \u003cem\u003ePF3D7_1035100\u003c/em\u003e (unknown function, Southeast Asia), and \u003cem\u003ePF3D7_1035300\u003c/em\u003e (\u003cem\u003eGLURP\u003c/em\u003e, Southeast Asia).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo complement signals of recent positive selection, we identified and characterised genomic regions enriched for such signals and annotated the associated gene products. A total of 41 unique genomic regions showed enrichment for recent positive selection (\u003cstrong\u003eSupplementary Table 11\u003c/strong\u003e), highlighting loci potentially shaped by local adaptation. These regions encoded gene products linked to erythrocyte invasion, drug resistance, virulence, and the regulation of gene expression, including epigenetic modification. Seven regions, East Africa, Central Africa, South Central Africa, Horn of Africa, West Africa, Oceania, and South Asia, shared a selection signal on chromosome 8 (1.300,000\u0026ndash;1,330,000), encoding the PHISTc protein family. Another prominent signal was observed on chromosome 4 (1,090,000\u0026ndash;1,120,000), present across East Africa, Central Africa, South Central Africa, West Africa, Oceania and South Asia. This region encodes the erythrocyte binding antigen-165 (EBA-165), reinforcing the role of erythrocyte invasion pathways as key targets of selection. Central Africa also exhibited strong positive selection across chromosome 7 (400,000\u0026ndash;430,000), which contains \u003cem\u003epfcrt\u003c/em\u003e, which is linked to chloroquine resistance. Additional regions of interest include chromosome 8 (1,350,000\u0026ndash;1,380,000), encoding \u003cem\u003eHSP70x\u003c/em\u003e, a protein involved in the export of virulence factors, and a region encoding \u003cem\u003eBDP4\u003c/em\u003e, an epigenetic regulator of gene expression observed under selection across multiple regions \u003csup\u003e42\u003c/sup\u003e. Furthermore, the detection of selection around \u003cem\u003epfubp1\u003c/em\u003e in East Africa is notable given emerging evidence linking this gene to antimalarial resistance \u003csup\u003e43\u003c/sup\u003e. Collectively, these findings highlight chromosomal hotspots where adaptive pressures, particularly those related to host-parasite interactions and drug pressure, have likely shaped \u003cem\u003eP. falciparum\u003c/em\u003e evolution.\u003c/p\u003e\n\u003cp\u003eBetween-population (XP-EHH) analyses revealed a total of 266 SNPs across 75 genes with evidence of divergent selection between regions (Threshold:\u0026gt;5) (\u003cstrong\u003eSupplementary Table 12\u003c/strong\u003e). The greatest number of differentiated sites were observed in comparisons involving Central Africa, particularly with East Africa (n = 152), West Africa (n = 140) and South Central Africa (n = 134). Several genes exhibited high numbers of XP-EHH signals across comparisons, including \u003cem\u003ePF3D7_0713000\u003c/em\u003e (RIF), \u003cem\u003ePF3D7_0709300\u003c/em\u003e (CG2), \u003cem\u003ePF3D7_0710200\u003c/em\u003e, \u003cem\u003ePF3D7_1475900\u003c/em\u003e, and \u003cem\u003ePF3D7_0113800\u003c/em\u003e, each containing over 100 significant sites. GO term analysis of genes with elevated XP-EHH signals identified significant fold changes for biological processes involved in interspecies interactions (P \u0026lt; 0.05; \u003cstrong\u003eSupplementary Table 13\u003c/strong\u003e). Notably, this included \u003cem\u003ePF3D7_0209000\u003c/em\u003e (\u003cem\u003eP230\u003c/em\u003e), which is essential for ookinete formation and mosquito transmission (Central vs. East and South Central Africa), and \u003cem\u003ePF3D7_1346800\u003c/em\u003e (\u003cem\u003eP47\u003c/em\u003e), which mediates mosquito immune evasion (East vs. West Africa and Central vs West Africa). These findings suggest that positive selection in different regions may be influenced by local adaptation to \u003cem\u003eAnopheles\u003c/em\u003e mosquito vectors.\u003c/p\u003e\n\u003cp\u003eFurther investigation of chromosomal regions enriched for XP-EHH signals revealed 39 distinct regions under differential selection across populations (\u003cstrong\u003eSupplementary Table 14\u003c/strong\u003e). Of particular interest were two regions associated with antimalarial resistance. These included a locus encoding \u003cem\u003ePPPK-DHPS\u003c/em\u003e on chromosome 8 (530,000\u0026ndash;550,000), associated with sulfadoxine resistance (Central vs. East or West Africa), and the \u003cem\u003epfcrt\u0026nbsp;\u003c/em\u003egene on chromosome 7 (400,000\u0026ndash;430,000; 360,000\u0026ndash;450,000), implicated in chloroquine resistance. XP-EHH signals at the \u003cem\u003epfcrt\u003c/em\u003e locus were detected in comparisons such as Central Africa vs. East Africa, Horn of Africa vs. South Central Africa and South Asia vs. South Central Africa, among others. These findings suggest that variation in historical drug treatment regimens may have driven region-specific selection pressures across \u003cem\u003eP. falciparum\u003c/em\u003e populations.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSignals of selection vary across countries over time\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo complement haplotype-based approaches, we also examined signatures of recent positive selection using the iR statistic, which identifies loci with excess IBD sharing. The top five genes with the strongest selection signals (\u0026minus;log₁₀P \u0026gt; 5) for each country and time period are reported (\u003cstrong\u003eSupplementary Figure 10, Supplementary Table 15\u003c/strong\u003e). Several genes consistently exhibited strong selection signals across multiple countries and time frames, including \u003cem\u003eABCK1\u003c/em\u003e, \u003cem\u003eAQP2\u003c/em\u003e, \u003cem\u003eARO\u003c/em\u003e, \u003cem\u003eATPase2\u003c/em\u003e, \u003cem\u003eCAF1\u003c/em\u003e, \u003cem\u003eCARM1\u003c/em\u003e, \u003cem\u003eCG2\u003c/em\u003e, \u003cem\u003eCRK3\u003c/em\u003e, \u003cem\u003epfcrt\u003c/em\u003e, \u003cem\u003eCUL1\u003c/em\u003e, \u003cem\u003eDHHC4\u003c/em\u003e, \u003cem\u003eDRN1\u003c/em\u003e, \u003cem\u003eJmjC1\u003c/em\u003e, \u003cem\u003eMC-2TM\u003c/em\u003e, and members of the \u003cem\u003eFIKK\u0026nbsp;\u003c/em\u003efamily. Among these, \u003cem\u003epfcrt\u003c/em\u003e and \u003cem\u003epfdhps\u003c/em\u003e (\u003cem\u003ePPPK-DHPS\u003c/em\u003e), particularly in Papua New Guinea (2012\u0026ndash;2015), were notable for their well-established roles in antimalarial drug resistance. In addition to these canonical resistance genes, \u003cem\u003eABCK1\u003c/em\u003e (an ATP-binding cassette transporter) and \u003cem\u003eMC-2TM\u003c/em\u003e (a multidrug and toxin extrusion transporter) also showed strong signals of selection, suggesting possible involvement in drug efflux mechanisms and warranting further functional investigation. Some genes showed evidence of recent positive selection in specific countries only, pointing to localised adaptation. These include \u003cem\u003eARF-GAP\u003c/em\u003e and \u003cem\u003eMCM3\u003c/em\u003e (The Gambia, 2016\u0026ndash;2019); \u003cem\u003eCDPK6\u003c/em\u003e, \u003cem\u003eCDPK7\u003c/em\u003e, \u003cem\u003eCK2\u0026alpha;\u003c/em\u003e, and \u003cem\u003eCLK3\u003c/em\u003e (Kenya, 2016\u0026ndash;2019); \u003cem\u003eDHHC9\u003c/em\u003e (Laos, 2016\u0026ndash;2019); \u003cem\u003eHAS1\u003c/em\u003e (Benin, 2016\u0026ndash;2019); \u003cem\u003eIMC1g\u003c/em\u003e (The Gambia, 2016\u0026ndash;2019); \u003cem\u003eLSA1\u003c/em\u003e and \u003cem\u003eM712\u0026nbsp;\u003c/em\u003e(Zambia, 2016\u0026ndash;2019); \u003cem\u003eNHE\u003c/em\u003e (Guinea, 2016\u0026ndash;2019); \u003cem\u003eP36\u003c/em\u003e (Papua New Guinea, 2016\u0026ndash;2019); \u003cem\u003ePF3D7_0201300\u0026nbsp;\u003c/em\u003e(Mozambique, 2016\u0026ndash;2019); and \u003cem\u003eRAB7\u003c/em\u003e (Guyana, 2020\u0026ndash;2021). These genes are involved in key cellular processes including metabolism, cell division, signal transduction, protein synthesis, membrane trafficking, and host-cell invasion. Several of the genes under country-specific selection, such as \u003cem\u003eLSA1\u003c/em\u003e (Liver Stage Antigen 1) and \u003cem\u003eP36\u003c/em\u003e, are potential malaria vaccine candidates. Evidence of selection at these loci may have implications for vaccine efficacy and highlights the need to monitor adaptive evolution in \u003cem\u003eP. falciparum\u003c/em\u003e populations across geographic regions and over time \u003csup\u003e44,45\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeviation from neutral evolution\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTajima\u0026rsquo;s D\u0026nbsp;is a widely used statistic for identifying regions of the genome that deviate from neutrality, offering insights into evolutionary processes such as\u0026nbsp;balancing selection\u0026nbsp;(characterised by an excess of intermediate-frequency variants; Tajima\u0026rsquo;s D \u0026gt; 2) and\u0026nbsp;selective sweeps\u0026nbsp;(marked by an excess of rare alleles; Tajima\u0026rsquo;s D \u0026lt; \u0026ndash;2) (\u003cstrong\u003eSupplementary Figure 11\u003c/strong\u003e). Relatively few 10 kb genomic windows were identified as being under balancing selection across all geographical regions, with the highest numbers observed in South America (n=16), Oceania (n=11), Southeast Asia (n=8), Horn of Africa (n=6), South Asia (n=4), Central Africa (n=2), and South Central Africa (n=1). These regions predominantly map to loci previously implicated in immune-mediated selection. In contrast, a larger number of genomic regions showed evidence of an excess of rare alleles, potentially indicative of recent or partial selective sweeps. The number of such regions varied across geography: Oceania (n=1,938), South America (n=1,797), Horn of Africa (n=1,751), Southeast Asia (n=1,416), South Asia (n=766), South Central Africa (n=337), Central Africa (n=289), East Africa (n=149), and West Africa (n=99). While many of these may represent false positives, as suggested by comparisons with complementary statistics like H12, some may reflect soft selective sweeps that evade detection by metrics such as iHS. A notable example is a region on chromosome 13 (1720000\u0026ndash;1820000), which includes the \u003cem\u003epfkelch13\u003c/em\u003e gene, known to contain markers of artemisinin resistance. This region displayed low Tajima\u0026rsquo;s D values across multiple populations: East Africa (\u0026ndash;2.60), West Africa (\u0026ndash;2.56), South Central Africa (\u0026ndash;2.54), Central Africa (\u0026ndash;2.49), Southeast Asia (\u0026ndash;2.28) and South Asia (\u0026ndash;2.24).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGeographical distribution of drug-resistance markers\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe first examined the geographical distribution of known genotypic markers associated with antimalarial drug resistance. Established drug resistance markers were obtained from the Malaria-Profiler database and the WHO watchlist for \u003cem\u003epfkelch13\u003c/em\u003e artemisinin resistance markers \u003csup\u003e20,46\u003c/sup\u003e . The highest combined prevalence of resistance to chloroquine, pyrimethamine, and sulfadoxine was observed in South America (69.5%), followed by South Asia (67.0%), West Africa (62.2%), Central Africa (61.3%), East Africa (57.2%), Oceania (57.1%), South Central Africa (46.3%) and the Horn of Africa (40.3%) (\u003cstrong\u003eFig. 3\u003c/strong\u003e). In contrast, Southeast Asia showed the highest proportion of genotypes resistant to artemisinin in combination with these three drugs (40.1%), reflecting the region\u0026rsquo;s long-standing issue with multidrug-resistant \u003cem\u003eP. falciparum\u003c/em\u003e. Given the global dependence on ACTs as the frontline treatment for \u003cem\u003eP. falciparum\u003c/em\u003e malaria, the emergence and spread of artemisinin resistance poses a major threat to malaria control efforts. While artemisinin resistance remains most prevalent in Southeast Asia, genotypic markers were also detected in Oceania (n = 2), South Central Africa (n = 1), and South America (n = 1). The most commonly detected artemisinin resistance\u0026ndash;associated mutation was \u003cem\u003epfkelch13\u003c/em\u003e C580Y, present in 2,270 isolates from Southeast Asia and in two samples from Oceania. Other notable \u003cem\u003epfkelch13\u003c/em\u003e variants observed in Southeast Asia included P441L (2%), R539T (2%), F446I (1%) and Y493H (1%) (\u003cstrong\u003eSupplementary Figure 12\u003c/strong\u003e). Outside Southeast Asia, a single isolate from Zambia (South Central Africa) carried the P441L mutation, which is linked to partial resistance, while the R561H variant, a validated marker of artemisinin resistance, was detected in an isolate from Guyana (South America). A mutation at the same codon as C580Y, C580F, was also identified in one isolate from Myanmar, suggesting continued diversification of resistance-associated alleles in the region. Other mutations associated with full and partial artemisinin resistance have been detected through genotyping and targeted sequencing efforts in countries such as Uganda (R561H, A675V), Rwanda (R561H), Tanzania (C469Y, R561H), Kenya (C469Y, P574L) and Ethiopia (F446I, R662I, P574L); however, these mutations were not observed in the present dataset for these regions \u003csup\u003e17,47-50\u003c/sup\u003e. We further searched for missense mutations in the propeller domain of Kelch13 (\u003cstrong\u003eSupplementary Figure 13\u003c/strong\u003e). Fourteen mutations were uniquely observed in West Africa (N458D, E509D, V534L, A557S, T573S, E596G, E612D, G665S, E691D, L722V, C532S, V637I, V566I and V589I), while three mutations were detected in East Africa (Y630F, I634L and S522C). Additionally, the \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003emutations Q613E and R622T were identified in Central Africa and South Central Africa, respectively. However, all were reported at low frequency (\u0026lt;5 samples total). The N458D mutation is located near N458Y, a validated marker of artemisinin resistance; however, N458D itself is not validated nor widely reported as being associated with resistance \u003csup\u003e51\u003c/sup\u003e. The \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003eA578S variant was observed in Central Africa (n=2), East Africa (n=5), South Asia (n=5), South Central Africa (n=1), and West Africa (n=21), but is not considered linked to drug resistance and is known to be commonly found across Africa \u003csup\u003e52\u003c/sup\u003e. The \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003eS522C mutation has been reported in West Africa (n=4) and East Africa (n=1); although it has been associated with delayed parasite clearance, it has not been designated a candidate marker by WHO due to limited supporting data \u003csup\u003e52,53\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIBD patterns and SNP associations with \u003cem\u003epfkelch13\u003c/em\u003e-mediated resistance\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur previous analyses did not identify strong signals of positive selection at the \u003cem\u003epfkelch13\u003c/em\u003e locus specifically; however, the spread of \u003cem\u003epfkelch13\u003c/em\u003e mutations may involve polygenic adaptation, which can obscure traditional signatures of selective sweeps \u003csup\u003e54\u003c/sup\u003e. To investigate this further, we performed IBD analyses to map transmission networks and characterise the genomic background associated with \u003cem\u003epfkelch13\u003c/em\u003e mutations, focusing on Southeast Asia and Oceania (\u003cstrong\u003eFig. 4\u003c/strong\u003e). Overall, \u003cem\u003epfkelch13\u003c/em\u003e C580Y-positive samples exhibited significantly higher pairwise IBD fractions (median: 0.276) compared to samples with other \u003cem\u003epfkelch13\u003c/em\u003e missense mutations in the propeller domain (median: 0.06; Wilcoxon, P \u0026lt; 2 \u0026times; 10⁻\u0026sup1;⁶) or wild-type alleles (median: 0.029; Wilcoxon, P \u0026lt; 2 \u0026times; 10⁻\u0026sup1;⁶) (\u003cstrong\u003eFig. 4\u003c/strong\u003e). These findings support the hypothesis of clonal expansion of the C580Y genotype, particularly in the context of the KEL1/PLA1 lineage, a multidrug-resistant strain associated with piperaquine resistance \u003csup\u003e55\u003c/sup\u003e. To gain further insight, IBD networks were constructed using thresholds of 47.5% and 95% IBD (\u003cstrong\u003eSupplementary Figures 14\u0026ndash;16\u003c/strong\u003e). These thresholds were chosen to reflect broader regional transmission (47.5%) and more direct, local transmission (95%). While the convenience sampling approach may limit the ability to fully reconstruct historical transmission patterns, data were stratified into three epidemiologically relevant time windows: 2008\u0026ndash;2011 (emergence of \u003cem\u003epfkelch13\u003c/em\u003e C580Y in western Cambodia), 2012\u0026ndash;2015 (expansion of artemisinin resistance and the rise of KEL1/PLA1) and 2016\u0026ndash;2019 (regional dominance of artemisinin resistance across Southeast Asia) (\u003cstrong\u003eSupplementary Figures 14\u0026ndash;16\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAt the 95% IBD threshold, samples consistently clustered by \u003cem\u003epfkelch13\u003c/em\u003e genotype across all time periods, indicating transmission of closely related parasites carrying the same resistance mutations. Clustering also occurred by country or site, suggesting either local transmission or the onward spread of imported cases. In contrast, at the 47.5% threshold, clusters frequently included a mix of \u003cem\u003epfkelch13\u003c/em\u003e genotypes, likely reflecting recombination over time among genetically related parasites sharing similar genomic backgrounds. One example is a large cluster from 2008\u0026ndash;2011 containing three genotypes (\u003cem\u003epfkelch13\u003c/em\u003e C580Y, \u003cem\u003epfkelch13\u003c/em\u003e 539T, and wild-type) across Cambodia, Vietnam and Laos (\u003cstrong\u003eSupplementary Figure 14\u003c/strong\u003e). Over time, both the number and geographic spread of \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003eC580Y-positive clusters increased, often spanning multiple countries. This trend likely reflects the expansion and regional dominance of the KEL1/PLA1 lineage, although definitive conclusions are limited by the convenience sampling design.\u003c/p\u003e\n\u003cp\u003eThe IBD analysis revealed clonal clusters of \u003cem\u003epfkelch13\u003c/em\u003e C580Y-positive parasites with extensive shared genomic regions, suggesting that additional loci may be under co-selection. To investigate this further, we performed genome-wide SNP association analyses that integrated co-occurrence patterns, genotypic correlations, and population-adjusted models to identify variants linked to the C580Y background across Southeast Asia (see \u003cstrong\u003eMethods\u003c/strong\u003e). Candidate mutations were prioritised into three confidence tiers: Tier 1 (high), Tier 2 (moderate) and Tier 3 (low, functional relevance-based) (see \u003cstrong\u003eMethods\u003c/strong\u003e). A total of 22 mutations across 20 genes met the Tier 1 criteria, showing strong support across multiple analyses, significant association (P \u0026lt; 0.0001), high linkage (R\u0026sup2; \u0026gt; 0.5), and large effect sizes (an odds ratio (OR) in the 95th percentile) \u003cstrong\u003e(Supplementary Table 16\u003c/strong\u003e). This set included \u003cem\u003eARPS10\u003c/em\u003e V127M, a previously reported interactor with \u003cem\u003epfkelch13\u003c/em\u003e C580Y, validating our approach \u003csup\u003e56\u003c/sup\u003e. Another top-ranked variant, \u003cem\u003eMyoF\u003c/em\u003e S969P (PF3D7_1226000, formerly \u003cem\u003eMyoC\u003c/em\u003e), a component of the K13-associated protein complex, further implicates changes in the parasite\u0026rsquo;s intracellular endocytosis architecture in facilitating resistance in the Greater Mekong sub-region. An additional 55 mutations across 53 genes were classified as Tier 2 (moderate confidence), including \u003cem\u003epfcrt\u003c/em\u003e, which has been linked to the \u003cem\u003epfkelch13\u003c/em\u003e C580Y background. This supports the idea that a pre-existing drug resistance background, such as chloroquine resistance, may have predisposed parasites to evolve artemisinin resistance. However, recent studies of Ugandan isolates with K13-mediated reduced artemisinin susceptibility indicate that these have evolved on a background of wild-type \u003cem\u003epfcrt\u0026nbsp;\u003c/em\u003eand complete susceptibility to chloroquine \u003csup\u003e21,22\u003c/sup\u003e. Tier 3 consisted of 20 mutations in 14 genes (\u003cem\u003eMDR2, DHFR-TS, UBP1, AP2-L, MDR1, PF3D7_0907200, MCA2, AP2-G, VPS51, CRT, MyoC/MyoF, PF3D7_1365800, PF3D7_0907200, PF3D7_1329500\u003c/em\u003e, and \u003cem\u003ePF3D7_1243400\u003c/em\u003e) (\u003cstrong\u003eSupplementary Table 16\u003c/strong\u003e). These mutations show weaker statistical associations but were retained due to their potential functional relevance based on prior studies.\u003c/p\u003e\n\u003cp\u003eTo investigate potential shared pathways, we constructed a protein\u0026ndash;protein interaction (PPI) network using all prioritised candidates (\u003cstrong\u003eFig. 4F\u003c/strong\u003e). Six clusters emerged, including one enriched for known drug resistance proteins (MDR1, MDR2, CRT, DHFR-TS) and associated genes (\u003cem\u003ePF3D7_0214600, PF3D7_0104300, PF3D7_1438500, OXA1, PF3D7_1017000, PF3D7_1331300\u003c/em\u003e). This suggests that \u003cem\u003ePF3D7_0214600\u003c/em\u003e may have contributed to the emergence of \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003eC580Y. Additional interactions were observed between \u003cem\u003epfkelch13\u003c/em\u003e and PF3D7_0303800, a Tier 2 candidate found in Laos, as well as among \u003cem\u003ePF3D7_0405400\u003c/em\u003e (putative pre-mRNA processing-splicing factor 8), \u003cem\u003ePF3D7_1119300\u003c/em\u003e (splicing factor U2AF small subunit), and a duplicate listing of \u003cem\u003ePF3D7_1119300\u003c/em\u003e, all of which are involved in mRNA splicing, potentially influencing gene regulation and parasite fitness. Together, these associations highlight a complex genomic background linked to \u003cem\u003epfkelch13\u003c/em\u003e C580Y. While not all identified mutations may be functionally relevant, some are likely to represent compensatory changes or modifiers that impact cellular pathways, gene expression or survival. Further functional validation is needed to delineate their roles in compensatory evolution and artemisinin resistance.\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eWith the increasing threat of \u003cem\u003eP. falciparum\u003c/em\u003e across the globe, substantial effort is required to achieve malaria elimination. Data-driven strategies and the development of new tools offer innovative approaches to aid the surveillance of \u003cem\u003eP. falciparum\u003c/em\u003e. However, generating such tools necessitates a comprehensive understanding of \u003cem\u003eP. falciparum\u003c/em\u003e on a global scale, which has recently become attainable through access to large-scale, publicly accessible data \u003csup\u003e26,31-35,37\u003c/sup\u003e. By delving further into the \u003cem\u003eP. falciparum\u003c/em\u003e genome, we can gain greater insight into the biological mechanisms that threaten malaria control and enhance our understanding of population genetics. This large-scale genomic data allows for a deeper exploration of \u003cem\u003eP. falciparum\u003c/em\u003e\u0026apos;s evolution, patterns of gene flow, and the impact of control efforts, offering critical information to guide malaria control strategies.\u003c/p\u003e\n\u003cp\u003eThe evolution of \u003cem\u003eP. falciparum\u003c/em\u003e has been dynamic, shaped significantly by human and vector migration, which has facilitated its global spread, and by control efforts, which have led to population reductions and bottlenecks \u003csup\u003e8\u003c/sup\u003e. Both historical migrations and recent interventions have left their marks, visible in the population structure analysis of \u003cem\u003eP. falciparum\u003c/em\u003e genome-wide SNPs \u003csup\u003e9\u003c/sup\u003e\u003cem\u003e.\u003c/em\u003e \u003cem\u003eP. falciparum\u003c/em\u003e isolates cluster according to continental boundaries, with some clustering according to specific geographical regions such as South Asia, Southeast Asia, and Oceania, while African samples are less distinct. The clustering of South American samples with African regions aligns with evidence suggesting multiple introductions of \u003cem\u003eP. falciparum\u003c/em\u003e from Africa, including during the transatlantic slave trade \u003csup\u003e36,57\u003c/sup\u003e. ADMIXTURE analysis further highlights the shared ancestry between intercontinental and intracontinental regions. For example, as mentioned, South American samples share ancestry with West African populations \u003csup\u003e57\u003c/sup\u003e. However, they also exhibit a distinct genetic signature, likely shaped by a population bottleneck during introduction and approximately 450 years of subsequent isolated evolution, local adaptation, or influence from ancestral lineages not represented in current West African samples. Consistent with multidimensional scaling analysis,\u0026nbsp;African populations exhibited overlapping ancestry, with samples from\u0026nbsp;East and Central Africa showing greater genetic similarity to each other\u0026nbsp;than to those from\u0026nbsp;West Africa. In contrast, the Horn of Africa forms a distinct genetic cluster, shaped by gene flow from African regions as well as some influence from Asia across the Indian Ocean and Arabian Sea. Outside Africa, there was\u0026nbsp;partial overlap in ancestry between Southeast Asia, Oceania and South Asia, reflecting shared evolutionary history or gene flow, although\u0026nbsp;each region retained distinct genetic signatures, indicative of region-specific selection or demographic events.\u003c/p\u003e\n\u003cp\u003eFurther exploration of the relationship between genetic and geographic structure was carried out using F\u003csub\u003eST\u003c/sub\u003e analysis. Generally, genetic differentiation correlated with geographic distance, except in regions with smaller sample sizes, for which genetic differentiation was only evident at intracontinental scale. Despite the high SNP diversity in African samples, they displayed the least genetic differentiation. This aligns with other analyses indicating a high degree of outcrossing within African populations, supported by higher MOI and lower IBD estimates. Genomic regions with high F\u003csub\u003eST\u0026nbsp;\u003c/sub\u003evalues provide insight into the drivers of genetic differentiation between geographical regions. Several genes with high F\u003csub\u003eST\u003c/sub\u003e sites are known to play roles in transmission, including the gene encoding P47, a protein used for mosquito immune evasion, as well as P48/P45, which is essential for ookinete formation and transmission \u003csup\u003e58-60\u003c/sup\u003e. These genes and their products are proposed targets for transmission-blocking vaccines. Moreover, highly differentiating sites were detected in drug-resistance genes, reflecting the administration of different drugs across various regions and time points. The widespread prevalence of drug-resistance mutations across all regions is concerning and underscores the importance of global surveillance. Sites with high F\u003csub\u003eST\u0026nbsp;\u003c/sub\u003emay be utilised for geographical classification and molecular barcoding to support such surveillance efforts.\u003c/p\u003e\n\u003cp\u003eDespite the dataset being curated from samples sequenced over several decades, the results from the population genetic analysis align broadly with current trends in malaria transmission and disease burden. High MOI, indicated by lower F\u003csub\u003eWS\u003c/sub\u003e values, was estimated for isolates from South Asia, West Africa, East Africa and Central Africa. This corresponds to the high transmission intensity and frequent outcrossing known to occur in these regions and recent global health reports \u003csup\u003e1,61,62\u003c/sup\u003e. Country-level analysis suggested increasing MOI in Mali and Kenya in recent years (2016 onwards), suggesting the sites sampled from these countries should be of key concern and targets for malaria intervention. Conversely, low MOI was reported in the Horn of Africa, South America, Oceania and Southeast Asia, which could indicate less outbreeding and lower transmission because of more effective malaria control measures. However, this may also be influenced by the smaller sample sizes from some of these regions. Specific countries might also exhibit low MOI due to geographic isolation, as previously observed in samples from the Bijag\u0026oacute;s islands \u003csup\u003e32\u003c/sup\u003e. These findings suggest that \u003cem\u003eP. falciparum\u0026nbsp;\u003c/em\u003einfections in South Asia and Southeast Asia should be considered independently during global assessments rather than being combined. This contrasts with observations made for \u003cem\u003eP. vivax\u003c/em\u003e whereby F\u003csub\u003eWS\u003c/sub\u003e values are marginally higher in South Asia compared to Southeast Asia, but the values are overall very similar, likely due to \u003cem\u003eP. vivax\u0026nbsp;\u003c/em\u003erelapse maintaining genetic diversity across regions with comparable transmission \u003csup\u003e63\u003c/sup\u003e. The MOI results are consistent with IBD analysis, where higher fractions were observed in the Horn of Africa, South America, Oceania and Southeast Asia, indicative of reduced outcrossing and lower transmission. High IBD fractions were particularly noted in genes involved in drug resistance. This is likely due to strong positive selection from intensive antimalarial drug administration. Under such selective pressure, IBD sharing at drug-resistance loci may increase, alongside neutral genomic regions linked to these genes \u003csup\u003e64\u003c/sup\u003e. This could explain the higher IBD fractions found near \u003cem\u003epfcrt,\u003c/em\u003e which contains chloroquine-resistance mutations. High fractions of IBD were observed in \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003einteraction candidates \u003cem\u003eKIC7\u003c/em\u003e and \u003cem\u003eKIC9\u0026nbsp;\u003c/em\u003eacross the Horn of Africa and South America, regions that may be affected by artemisinin-resistance \u003csup\u003e65\u003c/sup\u003e. This could indicate selection in these genes which should consequently be a target for future investigation into markers for artemisinin-resistance. In addition, it has been suggested that regions under strong positive selection can bias IBD analysis \u003csup\u003e64\u003c/sup\u003e. Further research is needed to mitigate this effect for downstream IBD-based inference, such as estimating effective population size.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMoreover, the impact of positive selection was evident through the assessment of regions with extended haplotype homozygosity within (iHS) and between (XP-EHH) populations. Across all geographical regions, SNPs in genes encoding cell surface proteins or those involved in host cell invasion, such as the SURFIN and CLAG families, exhibited strong signatures of positive selection. \u0026nbsp;The strong signals of positive selection in \u003cem\u003eSURFIN\u003c/em\u003e and \u003cem\u003eCLAG\u003c/em\u003e genes likely reflect recent or ongoing adaptive changes to overcome host immunity or improve invasion efficiency, which may coexist with balancing selection maintaining diversity at other sites or times \u003csup\u003e66,67\u003c/sup\u003e. When comparing selection between populations, mosquito interaction genes \u003cem\u003eP47\u0026nbsp;\u003c/em\u003eand \u003cem\u003eP230\u0026nbsp;\u003c/em\u003ewere found to have undergone cross-population selection, between regions and countries \u003csup\u003e58\u003c/sup\u003e. This could be underpinned by ecological dynamics. For example, in Southeast Asia, where multiple \u003cem\u003eAnopheles\u003c/em\u003e species coexist and the dominant species can vary \u003csup\u003e68,69\u003c/sup\u003e. Signals of positive selection were also detected in drug-resistance genes, such as \u003cem\u003epfcrt\u003c/em\u003e and \u003cem\u003ePPPK-DHPS\u003c/em\u003e. This was most likely driven by different treatment regimens implemented across regions at different time points. The role of loci involved in gene expression, such as \u003cem\u003eBDP4\u003c/em\u003e, should also not be ignored. The regulation of gene expression and epigenetic modification support the parasite\u0026rsquo;s complex lifecycle and provide an additional mechanism for adaptation \u003csup\u003e70\u003c/sup\u003e. Together, examining regions under positive selection offers valuable insights into the past, present and future evolution of \u003cem\u003eP. falciparum\u003c/em\u003e and can guide its control.\u003c/p\u003e\n\u003cp\u003eAfter detecting the influence of anti-malarial treatment across \u003cem\u003eP. falciparum\u003c/em\u003e genomes, we aimed to explore the patterns of drug-resistance mutations further. The geographic distribution of these mutations aligned with past and current drug administration programs in each country and region. For instance, \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003emutations were predominantly observed in samples from Southeast Asia and Oceania \u003csup\u003e17\u003c/sup\u003e. However, these mutations were also observed in South America and South Central Africa, demonstrating the greater need for surveillance of \u003cem\u003epfkelch13\u003c/em\u003e markers outside of Southeast Asia \u003csup\u003e18\u003c/sup\u003e. Profiling drug-resistant strains could be improved by considering drug-resistant haplotypes and exploring potential compensatory effects further, as well as examining structural variants and copy number variants that were not covered in this study \u003csup\u003e26\u003c/sup\u003e. Using a convenience sample approach is beneficial for examining large-scale drug-resistance data, but caution is necessary when interpreting results due to temporal changes that may occur.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe lack of strong positive selection detected in \u003cem\u003epfkelch13\u003c/em\u003e has indicated that there may be compensatory mechanisms driving the evolution of artemisinin resistance across Southeast Asia \u003csup\u003e56,71\u003c/sup\u003e. We highlight the clonal expansion of \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003eC580Y mutants using IBD analysis, while also showing that other mutations have independently arisen from similar genomic backgrounds. In the absence of phenotypic data, potential interactors with the most common \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003emutation (C580Y) were probed by leveraging the power of the large-scale genome-wide sequencing data to search for SNP linkage, co-occurrence and association. Some mutations may be \u0026lsquo;background mutations\u0026rsquo; associated with the genetic lineage of samples from Southeast Asia, which could be clarified through genome-wide association tests with the artemisinin drug susceptibility phenotype and the inclusion of population structure covariates \u003csup\u003e72\u003c/sup\u003e . Despite this, combined with knowledge of protein-protein interactions from the literature, several candidates were shortlisted as potential candidates, including \u003cem\u003epfcrt\u003c/em\u003e and \u003cem\u003eARPS10\u003c/em\u003e which have previously been reported \u003csup\u003e56\u003c/sup\u003e, as well as \u003cem\u003epfubp1,\u0026nbsp;\u003c/em\u003ewhich was reported as undergoing positive selection in East Africa in this study. Additional candidates include \u003cem\u003eMyoF\u0026nbsp;\u003c/em\u003eand \u003cem\u003eRAD5. MyoF\u0026nbsp;\u003c/em\u003eis associated with the K13 compartment; \u003cem\u003eRAD5\u0026nbsp;\u003c/em\u003eis also reported as being under selection in the Greater Mekong Sub-region, where the KEL1/PLA1 lineage is known to dominate\u0026nbsp;\u003csup\u003e73,74\u003c/sup\u003e. While further\u0026nbsp;validation is needed, the identification of new candidate mutations provides valuable insights into the potential drivers of artemisinin resistance that may emerge in other regions globally. \u0026nbsp;The majority of isolates in the curated dataset were collected before the significant emergence of African \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003epropeller domain variants, which were first reported around 2020\u0026nbsp;\u003csup\u003e17,47\u003c/sup\u003e. As such, further analysis of these variants is limited in this dataset and represents an important area for future research.\u003c/p\u003e\n\u003cp\u003eThis study has identified specific patterns of genetic variation in \u003cem\u003eP. falciparum\u0026nbsp;\u003c/em\u003eacross the globe which varied over time and between countries and regions. Such information, integrated with further validation and additional study can help to rapidly channel genomic surveillance efforts into prompt interventions, with particular focus on screening drug-resistant mutations, adaptation to the \u003cem\u003eAnopheles\u0026nbsp;\u003c/em\u003evector and geographical classification \u003csup\u003e20\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e"},{"header":"METHODS","content":"\u003cp\u003e\u003cstrong\u003eSequence Data and Pre-processing raw reads\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA combined dataset comprised of 23,462 \u003cem\u003eP. falciparum\u0026nbsp;\u003c/em\u003eWGS were considered for analysis, including 16,203 high quality samples which were obtained from the MalariaGen Pf7 data resource \u003csup\u003e26\u003c/sup\u003e, 7,183 previously published samples \u003csup\u003e31-35,37,64,75,76\u003c/sup\u003e and 76 newly sequenced isolates (Brazil and Vietnam) which were processed for this study using previously described methods \u003csup\u003e31\u003c/sup\u003e. \u0026nbsp;After pre-processing the raw genomic data, duplicate samples, mixed species and low-quality samples were removed from the dataset, leaving 17,565 high quality samples for further analysis (\u003cstrong\u003eSupplementary Table 17\u003c/strong\u003e). Raw paired end sequence data was processed using an established bioinformatic pipeline (https://github.com/LSHTMPathogenSeqLab/fastq2matrix). FastQ files were obtained from the European Nucleotide Archive and were trimmed to remove poor quality sequences using trimmomatic(v0.39) and the following parameters: LEADING:3 TRAILING:3 SLIDINGWINDOW:4:20 MINLEN:36\u0026nbsp;\u003csup\u003e77\u003c/sup\u003e. Trimmed sequences were subsequently aligned to the \u003cem\u003eP. falciparum\u0026nbsp;\u003c/em\u003e3D7 reference genome v3 (GCA_000002765.3) using bwa-mem (v0.7.17-r1188) to produce BAM files\u0026nbsp;\u003csup\u003e78\u003c/sup\u003e. Samtools v1.18 (fixmate and markdup) was used to correct mate information after mapping with bwa-mem\u0026nbsp;\u003csup\u003e79\u003c/sup\u003e. Base quality score recalibration (BQSR) and correction were then performed using GATK (v4.1.4.1) (BaseRecalibrator and ApplyBQSR) to reduce systematic errors in the quality score of base calls derived from the sequencing process\u0026nbsp;\u003csup\u003e80\u003c/sup\u003e. This was carried out using the \u003cem\u003eP. falciparum\u0026nbsp;\u003c/em\u003egenetic crosses 1.0 dataset (https://www.malariagen.net/data_package/pf-crosses-1-0/).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eVariant calling (SNPs and small Indels) were called by also using GATK (v4.1.4.1) software (HaplotypeCaller) to create per-sample gVCF files (parameters: -ERC GVCF) \u003csup\u003e81\u003c/sup\u003e. The GenomicsDB datastore was used to store validated VCFs via the GATK\u0026rsquo;s GenomicsDBImport function. A multi-sample VCF file was then created using the GenotypeGVCFs function and further quality score recalibration was carried out using the GATK Variant Quality Score Recalibration (VQSR) function (parameters: -an QD -an FS -an SOR -an DP -maxGaussians 8 and -mq-cap-for-logit-jitter-transform 70). Variant Quality Score Log-Odds (VQSLOD) were obtained using the ApplyVQSR function (parameter: -truth-sensitivity-filter-level 99.0, \u003cem\u003eP. falciparum\u0026nbsp;\u003c/em\u003egenetic crosses 1.0 dataset) and variants with VQSLOD score\u0026thinsp;\u0026lt;\u0026thinsp;0 were filtered out to retain high-quality variant calls. Only SNPs found within the core genome were retained in the dataset. Further quality metrics were used to ensure the remaining dataset was of high quality. Isolates with \u0026gt;40% missing data were removed, leaving high quality isolates for population genetic analysis that had consistent coverage across the core genome. The genotype of SNPs with mixed calls with a secondary MAF\u0026gt;20% were determined by the ratio of coverage. Variants were annotated using snpEff (v5.1). After all filtering steps, 17,565 high quality samples remained for further analysis. The high-quality multi-sample VCF was filtered for bi-allelic SNPs. Two VCF files were generated: one containing only bi-allelic SNPs, and a second containing normalized multi-allelic sites, where only the alternative allele with the highest minor allele frequency (MAF) was retained. The normalised VCF underwent additional filtering to remove additional hypervariable genes, which were defined as being in the top 95% quartile for SNP density (SNP per bp) (\u003cstrong\u003eSupplementary Table 18\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePopulation Genetic Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMultidimensional scaling (MDS) was carried out using all samples and SNPs. A distance matrix was first calculated using PLINK (v1.90) software using a filtered, bi-allelic VCF file \u003csup\u003e82\u003c/sup\u003e. The MDS was carried out over the distance matrix using R (v4.2.2). F\u003csub\u003eST\u003c/sub\u003e analysis and estimation of SNP diversity (\u0026pi;) was carried out using the python package scikit-allele (v1.37) (https://github.com/cggh/scikit-allel). The VCF file was converted to a ZARR format and F\u003csub\u003eST\u003c/sub\u003e analysis was performed between regions, as well as a \u0026lsquo;one-against-all\u0026rsquo; approach whereby samples from each region were compared to the rest of the dataset in the analysis. F\u003csub\u003eST\u003c/sub\u003e analysis was only carried out using segregating sites. Thresholds of F\u003csub\u003eST\u003c/sub\u003e \u0026gt;0.75 (moderate) and F\u003csub\u003eST\u003c/sub\u003e \u0026gt;0.95 (high) were used to identify sites with moderate-high F\u003csub\u003eST\u003c/sub\u003e. Correlation between genetic and geographic distances (F\u003csub\u003eST\u003c/sub\u003e) were estimated using the Mantel test using the vegan package in R (https://github.com/vegandevs/vegan). Mean pairwise difference was used to estimate SNP diversity in 10kb windows, excluding sites with \u0026gt;80% missingness and SNPs in non-coding regions. However, results from SNP diversity analyses are heavily influenced by missingness thresholds and lack of invariant sites so may not be reliable estimates and should be interpreted with caution. All code for these analyses is available on GitHub in a dedicated repository: https://github.com/NinaMercedes/PopGen/tree/main/Pop_Gen. Multiplicity of infection (MOI) was first estimated using only coding regions and bi-allelic SNPs. A population-specific MAF threshold of 1% was used to filter SNPs (threshold for all analyses unless stated otherwise). The multi-sample VCFs were split and filtered according to country and region using bcftools (v1.20) and the F\u003csub\u003eWS\u003c/sub\u003e metric was calculated using the moimix R package (v0.0.2.9001) (https://github.com/bahlolab/moimix). A F\u003csub\u003eWS\u003c/sub\u003e metric of \u0026gt;0.95 was used as a threshold to remove multiclonal isolates. F\u003csub\u003eWS\u003c/sub\u003e scores were compared across countries and year groups to identify changes F\u003csub\u003eWS\u003c/sub\u003e score using ANOVA. Year groups were defined on a distribution basis and spanned 3-year periods. P-values were adjusted using the false discovery rate and P\u0026lt;0.005 was used to indicate a significant difference between groups. \u0026nbsp;Downstream analysis was performed using biallelic SNPs only which were used to produce a binary genotype matrix (reference allele=0, alternative allele=1, mixed allele=0.5, missing allele=Ns). ADMIXTURE software (v1.3) (parameters: -cv=10 -j8 --haploid=\u0026quot;*\u0026quot;) was used to estimate individual ancestries across SNP genotypes\u0026nbsp;\u003csup\u003e83\u003c/sup\u003e. This was performed using a bed file which was converted from the multi-sample biallelic VCF file using PLINK (v1.90). Ten-fold cross-validation was used to estimate the optimal K value through inspection of the inflection point after running across K values 1 to 12. The optimal value was K=10 (\u003cstrong\u003eSupplementary Figure 17\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFurthermore, IBD sharing across genomes was estimated using hmmIBD software which utilised a hidden Markov chain model \u003csup\u003e84\u003c/sup\u003e. \u0026nbsp;The pairwise fraction collected for all specified regions and countries was used to estimate genomic relatedness. Genome-wide IBD fractions across 10kb sliding windows were also calculated. Regions were annotated with gene annotations and gene products for enhanced interpretation. Analysis of IBD at country and year group level were also performed using isoRelate and the iR statistic was used to identify sites undergoing positive selection \u003csup\u003e85\u003c/sup\u003e. The biallelic matrix was also used to determine regions under positive selection within integrated haplotype homozygosity score (Threshold: \u0026minus;log10[1\u0026minus;2|\u0026Phi;(iHS)-0.5|]\u0026gt;4.0) and between populations (XP-EHH) (Threshold: 5) using the rehh R package \u003csup\u003e86\u003c/sup\u003e. Highly variable regions such as \u003cem\u003evar\u0026nbsp;\u003c/em\u003egenes were removed from the analysis to prevent false positive results. The iHS is a within-population statistic, where a positive score indicates favourable selection for the ancestral allele and negative score indicates favourable selection for the alternative allele. The direction of the XP-EHH output corresponds to the population in which selection is taking place. Regions under positive selection were also annotated with gene annotations and gene products. Countries with fewer than 5 samples were excluded from population genetic analyses were applicable. Tajima\u0026rsquo;s D was estimated using VCFtools (v0.1.16) in 10kb windows. Genes of interest identified across regional level F\u003csub\u003eST\u003c/sub\u003e analyses (single and paired), IBD analysis, iHS and XPEHH, were annotated using GO terms (biological process, molecular function and cellular components) and KEGG pathways using PANTHER \u003csup\u003e87\u003c/sup\u003e. All selection analyses were also run using the normalised VCF file and binary matrix outputs. This was to prevent the exclusion of any important SNPs, including \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003eC580Yin selection analyses. However, this did not have a large overall impact on the overall results, except for the inclusion of additional variation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDrug-resistance mutations and \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003eC580Y associations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDrug-resistance mutations were assessed using the multi-allelic VCF file. Known drug-resistance mutations were retrieved from the malaria-profiler database (https://github.com/jodyphelan/malaria-db) and https://www.who.int/news-room/questions-and-answers/item/artemisinin-resistance which were used to filter the VCF files \u003csup\u003e20\u003c/sup\u003e. Additional nonsynonymous mutations were used by filtering gene boundaries using a bed file found within the database. This was used to generate a matrix for identifying additional mutations and nonsynonymous mutation combinations. Indels were also included in the matrix (PF3D7_0523000 N75E). Samples from Southeast Asia were subject to further IBD analysis. Outputs from hmmIBD were used to construct identity-by-descent (IBD) networks across Southeast Asia, across three-year time periods spanning from 2008 to 2019. Networks were visualised using ipysigma (https://github.com/medialab/ipysigma). 47.5% and 95% IBD thresholds were used to reveal broader, ancestral connectivity and recent, direct clonal expansion. Pairwise IBD fractions were compared across three genotypic groups: \u003cem\u003epfkelch13\u0026nbsp;\u003c/em\u003eC580Y, other missense mutations in the propeller domain, and wild-type. Statistical significance was assessed using the Kruskal-Wallis test for overall group differences, followed by two-sided Mann-Whitney U tests for pairwise comparisons.\u003c/p\u003e\n\u003cp\u003eA multi-layered association analysis was conducted to identify SNPs associated with the \u003cem\u003epfkelch13\u003c/em\u003e C580Y mutation using both a drug-resistance binary matrix and a genome-wide binary matrix filtered for common variants (MAF \u0026gt; 1%). Analyses were performed using two statistical frameworks: a logistic regression model adjusted for the top five principal components (across all samples, \u0026ldquo;All\u0026rdquo;), and latent factor models within each country (Cambodia, Laos, Vietnam, Myanmar, and Thailand), implemented in the LEA package to account for population structure and admixture (optimal K inferred via snmf) \u003csup\u003e88\u003c/sup\u003e. In both cases, P-values were adjusted using the Benjamini-Hochberg method. To complement the association models, we calculated an interaction coefficient (odds ratio, OR) was calculate, where OR = A\u0026times;D / B\u0026times;C based on co-occurrence of SNPs with C580Y (0 counts imputed as 0.5), and genotypic correlation (\u003cem\u003er\u0026sup2;\u003c/em\u003e) using VCFtools (v0.1.16). These four metrics, adjusted p-value from the logistic model, adjusted p-value from latent factor models, OR, and \u003cem\u003er\u0026sup2;\u003c/em\u003e were integrated to prioritise SNPs into low, medium and high confidence candidates (\u0026lsquo;Tiers\u0026rsquo;). Tier 1: SNPs meeting at least three of the four criteria (Adjusted P\u0026lt;0.001, r\u0026sup2; \u0026gt; 0.5, OR \u0026gt; 123). Tier 2: SNPs meeting two criteria, with moderate support (r\u0026sup2; \u0026gt; 0.2 and OR \u0026gt; 5). Tier 3: SNPs meeting one criterion and located in a curated list of genes with biological relevance or prior evidence (\u003cstrong\u003eSupplementary Table 19\u003c/strong\u003e) \u003csup\u003e74\u003c/sup\u003e. STRING protein-protein interactions (clustered using the k-means algorithm) was used to identify functional links between gene candidates \u003csup\u003e89\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAUTHORS CONTRIBUTIONS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSC and TGC conceived and supervised the project. JGD, LV, SM, JG, CJS, CRFM, NTHN, NTHB, and NQT conducted sample processing and DNA extraction. JGD, LV, SM, and SC conducted sequencing. JT provided software tools. JGD, LV, SM, JG, CJS, CRFM, NTHN, NTHB, and NQT provided data. NB performed bioinformatics and statistical analyses under the supervision of SC and TGC. All authors contributed to data interpretation. NB drafted the manuscript, with input from all authors. All authors reviewed, edited, and approved the final manuscript. NB, SC, and TGC compiled the final submission.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDECLARATION OF INTERESTS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAuthors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDATA SHARING\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRaw sequencing data are available in the European Nucleotide Archive (ENA). A complete list of accession numbers is provided in PRJEB94034.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eACKNOWLEDGEMENTS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was funded by MRC Newton UK \u0026ndash; MOST Vietnam (ref. MR/R026297/1) award. CRFM was supported by the S\u0026atilde;o Paulo Research Foundation-FAPESP (www.fapesp.br) [grant numbers 2024/10186-9] and the National Council for Scientific and Technological Development-CNPq (www.cnpq.br) [grant number 307193/2023-3]. JGD was supported by fellowships from FAPESP [grant numbers 2019/12068-5 and 2022/02771-3]. TGC and SC are funded by the UKRI (BBSRC BB/X018156/1; MRC MR/X005895/1; EPSRC EP/Y018842/1). The funders had no role in study design, data collection and analysis, decision to publish, or the preparation of the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eGeneva World Health Organization. World malaria report 2024: addressing inequity in the global malaria response. Geneva: World Health Organization; 2024. Licence: CC BY-NC-SA 3.0 IGO. (2024). \u003c/li\u003e\n\u003cli\u003eMasserey, T.\u003cem\u003e et al.\u003c/em\u003e The influence of biological, epidemiological, and treatment factors on the establishment and spread of drug-resistant Plasmodium falciparum. \u003cem\u003eeLife\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, e77634 (2022). https://doi.org/10.7554/eLife.77634\u003c/li\u003e\n\u003cli\u003eAdams, K. L.\u003cem\u003e et al.\u003c/em\u003e Selection for insecticide resistance can promote Plasmodium falciparum infection in Anopheles. \u003cem\u003ePLOS Pathogens\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, e1011448 (2023). https://doi.org/10.1371/journal.ppat.1011448\u003c/li\u003e\n\u003cli\u003eLeal Filho, W., May, J., May, M. \u0026amp; Nagy, G. J. Climate change and malaria: some recent trends of malaria incidence rates and average annual temperature in selected sub-Saharan African countries from 2000 to 2018. \u003cem\u003eMalaria Journal\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 248 (2023). https://doi.org/10.1186/s12936-023-04682-4\u003c/li\u003e\n\u003cli\u003eYalcindag, E.\u003cem\u003e et al.\u003c/em\u003e Multiple independent introductions of Plasmodium falciparum in South America. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e109\u003c/strong\u003e, 511-516 (2012). https://doi.org/10.1073/pnas.1119058109\u003c/li\u003e\n\u003cli\u003eTanabe, K.\u003cem\u003e et al.\u003c/em\u003e Plasmodium falciparum accompanied the human expansion out of Africa. \u003cem\u003eCurr Biol\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 1283-1289 (2010). https://doi.org/10.1016/j.cub.2010.05.053\u003c/li\u003e\n\u003cli\u003eTanabe, K.\u003cem\u003e et al.\u003c/em\u003e Plasmodium falciparum mitochondrial genetic diversity exhibits isolation-by-distance patterns supporting a sub-Saharan African origin. \u003cem\u003eMitochondrion\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 630-636 (2013). https://doi.org/10.1016/j.mito.2013.08.008\u003c/li\u003e\n\u003cli\u003eTagliamonte, M. S.\u003cem\u003e et al.\u003c/em\u003e Genetic Markers of Adaptation of Plasmodium falciparum to Transmission by American Vectors Identified in the Genomes of Parasites from Haiti and South America. \u003cem\u003emSphere\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e (2020). https://doi.org/10.1128/mSphere.00937-20\u003c/li\u003e\n\u003cli\u003eRodrigues, P. T.\u003cem\u003e et al.\u003c/em\u003e Human migration and the spread of malaria parasites to the New World. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 1993 (2018). https://doi.org/10.1038/s41598-018-19554-0\u003c/li\u003e\n\u003cli\u003eAnderson, T. J.\u003cem\u003e et al.\u003c/em\u003e Microsatellite markers reveal a spectrum of population structures in the malaria parasite Plasmodium falciparum. \u003cem\u003eMol Biol Evol\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 1467-1482 (2000). https://doi.org/10.1093/oxfordjournals.molbev.a026247\u003c/li\u003e\n\u003cli\u003eMachado, R. L.\u003cem\u003e et al.\u003c/em\u003e Genetic structure of Plasmodium falciparum populations in the Brazilian Amazon region. \u003cem\u003eJ Infect Dis\u003c/em\u003e \u003cstrong\u003e190\u003c/strong\u003e, 1547-1555 (2004). https://doi.org/10.1086/424601\u003c/li\u003e\n\u003cli\u003eIwagami, M.\u003cem\u003e et al.\u003c/em\u003e Genetic diversity and population structure of Plasmodium falciparum in the Philippines. \u003cem\u003eMalar J\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 96 (2009). https://doi.org/10.1186/1475-2875-8-96\u003c/li\u003e\n\u003cli\u003eMobegi, V. A.\u003cem\u003e et al.\u003c/em\u003e Population genetic structure of Plasmodium falciparum across a region of diverse endemicity in West Africa. \u003cem\u003eMalar J\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 223 (2012). https://doi.org/10.1186/1475-2875-11-223\u003c/li\u003e\n\u003cli\u003eSirawaraporn, W., Sathitkul, T., Sirawaraporn, R., Yuthavong, Y. \u0026amp; Santi, D. V. Antifolate-resistant mutants of Plasmodium falciparum dihydrofolate reductase. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e94\u003c/strong\u003e, 1124-1129 (1997). https://doi.org/10.1073/pnas.94.4.1124\u003c/li\u003e\n\u003cli\u003eStaedke, S. G.\u003cem\u003e et al.\u003c/em\u003e Relationship between age, molecular markers, and response to sulphadoxine-pyrimethamine treatment in Kampala, Uganda. \u003cem\u003eTrop Med Int Health\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 624-629 (2004). https://doi.org/10.1111/j.1365-3156.2004.01239.x\u003c/li\u003e\n\u003cli\u003eKublin, J. G.\u003cem\u003e et al.\u003c/em\u003e Molecular markers for failure of sulfadoxine-pyrimethamine and chlorproguanil-dapsone treatment of Plasmodium falciparum malaria. \u003cem\u003eJ Infect Dis\u003c/em\u003e \u003cstrong\u003e185\u003c/strong\u003e, 380-388 (2002). https://doi.org/10.1086/338566\u003c/li\u003e\n\u003cli\u003eUwimana, A.\u003cem\u003e et al.\u003c/em\u003e Emergence and clonal expansion of in vitro artemisinin-resistant Plasmodium falciparum kelch13 R561H mutant parasites in Rwanda. \u003cem\u003eNat Med\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 1602-1608 (2020). https://doi.org/10.1038/s41591-020-1005-2\u003c/li\u003e\n\u003cli\u003eNjiro, B. J.\u003cem\u003e et al.\u003c/em\u003e Molecular surveillance of chloroquine-resistant Plasmodium falciparum in sub-Saharan African countries after withdrawal of chloroquine for treatment of uncomplicated malaria: A systematic review. \u003cem\u003eJ Infect Public Health\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 550-557 (2022). https://doi.org/10.1016/j.jiph.2022.03.015\u003c/li\u003e\n\u003cli\u003eRavenhall, M.\u003cem\u003e et al.\u003c/em\u003e An analysis of large structural variation in global Plasmodium falciparum isolates identifies a novel duplication of the chloroquine resistance associated gene. \u003cem\u003eScientific Reports\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 8287 (2019). https://doi.org/10.1038/s41598-019-44599-0\u003c/li\u003e\n\u003cli\u003ePhelan, J. E.\u003cem\u003e et al.\u003c/em\u003e Rapid profiling of Plasmodium parasites from genome sequences to assist malaria control. \u003cem\u003eGenome Med\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 96 (2023). https://doi.org/10.1186/s13073-023-01247-7\u003c/li\u003e\n\u003cli\u003evan Schalkwyk, D. A.\u003cem\u003e et al.\u003c/em\u003e Treatment Failure in a UK Malaria Patient Harboring Genetically Variant Plasmodium falciparum From Uganda With Reduced In Vitro Susceptibility to Artemisinin and Lumefantrine. \u003cem\u003eClin Infect Dis\u003c/em\u003e \u003cstrong\u003e78\u003c/strong\u003e, 445-452 (2024). https://doi.org/10.1093/cid/ciad724\u003c/li\u003e\n\u003cli\u003eTumwebaze, P. K.\u003cem\u003e et al.\u003c/em\u003e Decreased susceptibility of Plasmodium falciparum to both dihydroartemisinin and lumefantrine in northern Uganda. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 6353 (2022). https://doi.org/10.1038/s41467-022-33873-x\u003c/li\u003e\n\u003cli\u003eVolkman, S. K.\u003cem\u003e et al.\u003c/em\u003e A genome-wide map of diversity in Plasmodium falciparum. \u003cem\u003eNature Genetics\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, 113-119 (2007). https://doi.org/10.1038/ng1930\u003c/li\u003e\n\u003cli\u003eSabeti, P. C.\u003cem\u003e et al.\u003c/em\u003e Detecting recent positive selection in the human genome from haplotype structure. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e419\u003c/strong\u003e, 832-837 (2002). https://doi.org/10.1038/nature01140\u003c/li\u003e\n\u003cli\u003eNeafsey, D. E., Taylor, A. R. \u0026amp; MacInnis, B. L. Advances and opportunities in malaria population genomics. \u003cem\u003eNat Rev Genet\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 502-517 (2021). https://doi.org/10.1038/s41576-021-00349-5\u003c/li\u003e\n\u003cli\u003eAbdel Hamid, M. M.\u003cem\u003e et al.\u003c/em\u003e Pf7: an open dataset of Plasmodium falciparum genome variation in 20,000 worldwide samples. \u003cem\u003eWellcome Open Res\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 22 (2023). https://doi.org/10.12688/wellcomeopenres.18681.1\u003c/li\u003e\n\u003cli\u003eAbera, D.\u003cem\u003e et al.\u003c/em\u003e Genomic analysis reveals independent evolution of Plasmodium falciparum populations in Ethiopia. \u003cem\u003eMalar J\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 129 (2021). https://doi.org/10.1186/s12936-021-03660-y\u003c/li\u003e\n\u003cli\u003eMwesigwa, A.\u003cem\u003e et al.\u003c/em\u003e Genetic diversity and population structure of Plasmodium falciparum across areas of varied malaria transmission intensities in Uganda. \u003cem\u003eMalar J\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 97 (2025). https://doi.org/10.1186/s12936-025-05325-6\u003c/li\u003e\n\u003cli\u003eCoulibaly, A.\u003cem\u003e et al.\u003c/em\u003e Genome-wide SNP analysis of Plasmodium falciparum shows differentiation at drug-resistance-associated loci among malaria transmission settings in southern Mali. \u003cem\u003eFront Genet\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 943445 (2022). https://doi.org/10.3389/fgene.2022.943445\u003c/li\u003e\n\u003cli\u003eNeafsey Daniel, E.\u003cem\u003e et al.\u003c/em\u003e Genetic Diversity and Protective Efficacy of the RTS,S/AS01 Malaria Vaccine. \u003cem\u003eNew England Journal of Medicine\u003c/em\u003e \u003cstrong\u003e373\u003c/strong\u003e, 2025-2037 (2015). https://doi.org/10.1056/NEJMoa1505819\u003c/li\u003e\n\u003cli\u003eVanheer, L. N.\u003cem\u003e et al.\u003c/em\u003e Genome-wide genetic variation and molecular surveillance of drug resistance in Plasmodium falciparum isolates from asymptomatic individuals in Ou\u0026eacute;less\u0026eacute;bougou, Mali. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 9522 (2023). https://doi.org/10.1038/s41598-023-36002-w\u003c/li\u003e\n\u003cli\u003eMoss, S.\u003cem\u003e et al.\u003c/em\u003e Population dynamics and drug resistance mutations in Plasmodium falciparum on the Bijag\u0026oacute;s Archipelago, Guinea-Bissau. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 6311 (2023). https://doi.org/10.1038/s41598-023-33176-1\u003c/li\u003e\n\u003cli\u003eOsborne, A.\u003cem\u003e et al.\u003c/em\u003e Plasmodium falciparum population dynamics in East Africa and genomic surveillance along the Kenya-Uganda border. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 18051 (2024). https://doi.org/10.1038/s41598-024-67623-4\u003c/li\u003e\n\u003cli\u003eFola, A. A.\u003cem\u003e et al.\u003c/em\u003e Genomics reveals heterogeneous Plasmodium falciparum transmission and selection signals in Zambia. \u003cem\u003eCommunications Medicine\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 67 (2024). https://doi.org/10.1038/s43856-024-00498-8\u003c/li\u003e\n\u003cli\u003eFlorimond, C.\u003cem\u003e et al.\u003c/em\u003e Impact of piperaquine resistance in Plasmodium falciparum on malaria treatment effectiveness in The Guianas: a descriptive epidemiological study. \u003cem\u003eThe Lancet Infectious Diseases\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 161-171 (2024). https://doi.org/https://doi.org/10.1016/S1473-3099(23)00502-9\u003c/li\u003e\n\u003cli\u003eLefebvre, M. J. M.\u003cem\u003e et al.\u003c/em\u003e Population Genomic Evidence of Adaptive Response during the Invasion History of Plasmodium falciparum in the Americas. \u003cem\u003eMolecular Biology and Evolution\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, msad082 (2023). https://doi.org/10.1093/molbev/msad082\u003c/li\u003e\n\u003cli\u003eVanhove, M.\u003cem\u003e et al.\u003c/em\u003e Temporal and spatial dynamics of Plasmodium falciparum clonal lineages in Guyana. \u003cem\u003ePLoS Pathog\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, e1012013 (2024). https://doi.org/10.1371/journal.ppat.1012013\u003c/li\u003e\n\u003cli\u003eOtto, T. D.\u003cem\u003e et al.\u003c/em\u003e Genomes of all known members of a Plasmodium subgenus reveal paths to virulent human malaria. \u003cem\u003eNature Microbiology\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 687-697 (2018). https://doi.org/10.1038/s41564-018-0162-2\u003c/li\u003e\n\u003cli\u003eBopp, S.\u003cem\u003e et al.\u003c/em\u003e Potent acyl-CoA synthetase 10 inhibitors kill Plasmodium falciparum by disrupting triglyceride formation. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 1455 (2023). https://doi.org/10.1038/s41467-023-36921-2\u003c/li\u003e\n\u003cli\u003eKeleta, Y., Ramelow, J., Cui, L. \u0026amp; Li, J. Molecular interactions between parasite and mosquito during midgut invasion as targets to block malaria transmission. \u003cem\u003enpj Vaccines\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 140 (2021). https://doi.org/10.1038/s41541-021-00401-9\u003c/li\u003e\n\u003cli\u003eKats, L. M.\u003cem\u003e et al.\u003c/em\u003e An exported kinase (FIKK4.2) that mediates virulence-associated changes in Plasmodium falciparum-infected red blood cells. \u003cem\u003eInt J Parasitol\u003c/em\u003e \u003cstrong\u003e44\u003c/strong\u003e, 319-328 (2014). https://doi.org/10.1016/j.ijpara.2014.01.003\u003c/li\u003e\n\u003cli\u003eDay, J., Passecker, A., Beck, H. P. \u0026amp; Vakonakis, I. The Plasmodium falciparum Hsp70-x chaperone assists the heat stress response of the malaria parasite. \u003cem\u003eFaseb j\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 14611-14624 (2019). https://doi.org/10.1096/fj.201901741R\u003c/li\u003e\n\u003cli\u003eXu, R.\u003cem\u003e et al.\u003c/em\u003e Deaggregation of mutant Plasmodium yoelii de-ubiquitinase UBP1 alters MDR1 localization to confer multidrug resistance. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1774 (2024). https://doi.org/10.1038/s41467-024-46006-3\u003c/li\u003e\n\u003cli\u003eCummings, J. F.\u003cem\u003e et al.\u003c/em\u003e Recombinant Liver Stage Antigen-1 (LSA-1) formulated with AS01 or AS02 is safe, elicits high titer antibody and induces IFN-gamma/IL-2 CD4+ T cells but does not protect against experimental Plasmodium falciparum infection. \u003cem\u003eVaccine\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 5135-5144 (2010). https://doi.org/10.1016/j.vaccine.2009.08.046\u003c/li\u003e\n\u003cli\u003eYadav, N.\u003cem\u003e et al.\u003c/em\u003e Identifying Plasmodium P36 and P52 antigens for coadministration with circumsporozoite protein to enhance vaccine efficacy. \u003cem\u003enpj Vaccines\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 241 (2024). https://doi.org/10.1038/s41541-024-01040-6\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. Malaria: Artemisinin partial resistance. https://www.who.int/news-room/questions-and-answers/item/artemisinin-resistance. (2025). \u003c/li\u003e\n\u003cli\u003eBalikagala, B.\u003cem\u003e et al.\u003c/em\u003e Evidence of Artemisinin-Resistant Malaria in Africa. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e385\u003c/strong\u003e, 1163-1171 (2021). https://doi.org/10.1056/NEJMoa2101746\u003c/li\u003e\n\u003cli\u003eJuliano, J. J.\u003cem\u003e et al.\u003c/em\u003e Prevalence of mutations associated with artemisinin partial resistance and sulfadoxine-pyrimethamine resistance in 13 regions in Tanzania in 2021: a cross-sectional survey. \u003cem\u003eLancet Microbe\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 100920 (2024). https://doi.org/10.1016/s2666-5247(24)00160-5\u003c/li\u003e\n\u003cli\u003eOsoti, V.\u003cem\u003e et al.\u003c/em\u003e Detection of low frequency artemisinin resistance mutations, C469Y, P553L and A675V, and fixed antifolate resistance mutations in asymptomatic primary school children in Kenya. \u003cem\u003eBMC Infectious Diseases\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 73 (2025). https://doi.org/10.1186/s12879-025-10462-z\u003c/li\u003e\n\u003cli\u003eAlemayehu, A. A.\u003cem\u003e et al.\u003c/em\u003e Expansion of the Plasmodium falciparum Kelch13 R622I Mutation in Northwest Ethiopia. \u003cem\u003eOpen Forum Infectious Diseases\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, ofaf279 (2025). https://doi.org/10.1093/ofid/ofaf279\u003c/li\u003e\n\u003cli\u003eTornyigah, B.\u003cem\u003e et al.\u003c/em\u003e Effect of Drug Pressure on Promoting the Emergence of Antimalarial-Resistant Parasites among Pregnant Women in Ghana. \u003cem\u003eAntimicrobial Agents and Chemotherapy\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, 10.1128/aac.02029-02019 (2020). https://doi.org/10.1128/aac.02029-19\u003c/li\u003e\n\u003cli\u003eSchmedes, S. E.\u003cem\u003e et al.\u003c/em\u003e Plasmodium falciparum kelch 13 Mutations, 9 Countries in Africa, 2014-2018. \u003cem\u003eEmerg Infect Dis\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 1902-1908 (2021). https://doi.org/10.3201/eid2707.203230\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. World Health Organization. Artemisinin resistance and artemisinin-based combination therapy efficacy: status report. 2018. (2018). \u003c/li\u003e\n\u003cli\u003ePritchard, J. K., Pickrell, J. K. \u0026amp; Coop, G. The Genetics of Human Adaptation: Hard Sweeps, Soft Sweeps, and Polygenic Adaptation. \u003cem\u003eCurrent Biology\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, R208-R215 (2010). https://doi.org/10.1016/j.cub.2009.11.055\u003c/li\u003e\n\u003cli\u003eAmato, R.\u003cem\u003e et al.\u003c/em\u003e Origins of the current outbreak of multidrug-resistant malaria in southeast Asia: a retrospective genetic study. \u003cem\u003eLancet Infect Dis\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 337-345 (2018). https://doi.org/10.1016/s1473-3099(18)30068-9\u003c/li\u003e\n\u003cli\u003eMiotto, O.\u003cem\u003e et al.\u003c/em\u003e Genetic architecture of artemisinin-resistant Plasmodium falciparum. \u003cem\u003eNature Genetics\u003c/em\u003e \u003cstrong\u003e47\u003c/strong\u003e, 226-234 (2015). https://doi.org/10.1038/ng.3189\u003c/li\u003e\n\u003cli\u003eMichel, M.\u003cem\u003e et al.\u003c/em\u003e Ancient Plasmodium genomes shed light on the history of human malaria. \u003cem\u003eNature\u003c/em\u003e (2024). https://doi.org/10.1038/s41586-024-07546-2\u003c/li\u003e\n\u003cli\u003eMolina-Cruz, A., Canepa, G. E. \u0026amp; Barillas-Mury, C. Plasmodium P47: a key gene for malaria transmission by mosquito vectors. \u003cem\u003eCurr Opin Microbiol\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 168-174 (2017). https://doi.org/10.1016/j.mib.2017.11.029\u003c/li\u003e\n\u003cli\u003eMolina-Cruz, A.\u003cem\u003e et al.\u003c/em\u003e Plasmodium falciparum evades immunity of anopheline mosquitoes by interacting with a Pfs47 midgut receptor. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e117\u003c/strong\u003e, 2597-2605 (2020). https://doi.org/10.1073/pnas.1917042117\u003c/li\u003e\n\u003cli\u003eMarin-Mogollon, C.\u003cem\u003e et al.\u003c/em\u003e The Plasmodium falciparum male gametocyte protein P230p, a paralog of P230, is vital for ookinete formation and mosquito transmission. \u003cem\u003eScientific Reports\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 14902 (2018). https://doi.org/10.1038/s41598-018-33236-x\u003c/li\u003e\n\u003cli\u003eAuburn, S.\u003cem\u003e et al.\u003c/em\u003e Characterization of Within-Host Plasmodium falciparum Diversity Using Next-Generation Sequence Data. \u003cem\u003ePLOS ONE\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, e32891 (2012). https://doi.org/10.1371/journal.pone.0032891\u003c/li\u003e\n\u003cli\u003eManske, M.\u003cem\u003e et al.\u003c/em\u003e Analysis of Plasmodium falciparum diversity in natural infections by deep sequencing. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e487\u003c/strong\u003e, 375-379 (2012). https://doi.org/10.1038/nature11174\u003c/li\u003e\n\u003cli\u003eNgwana-Joseph, G. C.\u003cem\u003e et al.\u003c/em\u003e Genomic analysis of global Plasmodium vivax populations reveals insights into the evolution of drug resistance. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 10771 (2024). https://doi.org/10.1038/s41467-024-54964-x\u003c/li\u003e\n\u003cli\u003eGuo, B.\u003cem\u003e et al.\u003c/em\u003e Strong positive selection biases identity-by-descent-based inferences of recent demography and population structure in Plasmodium falciparum. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 2499 (2024). https://doi.org/10.1038/s41467-024-46659-0\u003c/li\u003e\n\u003cli\u003eBirnbaum, J.\u003cem\u003e et al.\u003c/em\u003e A Kelch13-defined endocytosis pathway mediates artemisinin resistance in malaria parasites. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e367\u003c/strong\u003e, 51-59 (2020). https://doi.org/10.1126/science.aax4735\u003c/li\u003e\n\u003cli\u003eWinter, G.\u003cem\u003e et al.\u003c/em\u003e SURFIN is a polymorphic antigen expressed on Plasmodium falciparum merozoites and infected erythrocytes. \u003cem\u003eJournal of Experimental Medicine\u003c/em\u003e \u003cstrong\u003e201\u003c/strong\u003e, 1853-1863 (2005). https://doi.org/10.1084/jem.20041392\u003c/li\u003e\n\u003cli\u003eTrenholme, K. R.\u003cem\u003e et al.\u003c/em\u003e clag9: A cytoadherence gene in Plasmodium falciparum essential for binding of parasitized erythrocytes to CD36. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e97\u003c/strong\u003e, 4029-4033 (2000). https://doi.org/10.1073/pnas.040561197\u003c/li\u003e\n\u003cli\u003eSukkanon, C.\u003cem\u003e et al.\u003c/em\u003e Prevalence of Plasmodium spp. in Anopheles mosquitoes in Thailand: a systematic review and meta-analysis. \u003cem\u003eParasit Vectors\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 285 (2022). https://doi.org/10.1186/s13071-022-05397-2\u003c/li\u003e\n\u003cli\u003eSinka, M. E.\u003cem\u003e et al.\u003c/em\u003e A global map of dominant malaria vectors. \u003cem\u003eParasit Vectors\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 69 (2012). https://doi.org/10.1186/1756-3305-5-69\u003c/li\u003e\n\u003cli\u003eRead, D. F., Cook, K., Lu, Y. Y., Le Roch, K. G. \u0026amp; Noble, W. S. Predicting gene expression in the human malaria parasite Plasmodium falciparum using histone modification, nucleosome positioning, and 3D localization features. \u003cem\u003ePLOS Computational Biology\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, e1007329 (2019). https://doi.org/10.1371/journal.pcbi.1007329\u003c/li\u003e\n\u003cli\u003eMalaria, G. E. N. P. f. C. P. Genomic epidemiology of artemisinin resistant malaria. \u003cem\u003eeLife\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, e08714 (2016). https://doi.org/10.7554/eLife.08714\u003c/li\u003e\n\u003cli\u003eMiotto, P.\u003cem\u003e et al.\u003c/em\u003e A standardised method for interpreting the association between mutations and phenotypic drug resistance in Mycobacterium tuberculosis. \u003cem\u003eEur Respir J\u003c/em\u003e \u003cstrong\u003e50\u003c/strong\u003e (2017). https://doi.org/10.1183/13993003.01354-2017\u003c/li\u003e\n\u003cli\u003eNayak, S.\u003cem\u003e et al.\u003c/em\u003e Population genomics and transcriptomics of Plasmodium falciparum in Cambodia and Vietnam uncover key components of the artemisinin resistance genetic background. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 10625 (2024). https://doi.org/10.1038/s41467-024-54915-6\u003c/li\u003e\n\u003cli\u003eSchmidt, S.\u003cem\u003e et al.\u003c/em\u003e The Kelch13 compartment contains highly divergent vesicle trafficking proteins in malaria parasites. \u003cem\u003ePLoS Pathog\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, e1011814 (2023). https://doi.org/10.1371/journal.ppat.1011814\u003c/li\u003e\n\u003cli\u003e\u0026Ntilde;acata, I., Early, A. M., Boboy, J., Neafsey, D. E. \u0026amp; S\u0026aacute;enz, F. E. Effects of drug pressure and human migration on antimalarial resistance in circulating Plasmodium falciparum malaria parasites in Ecuador. \u003cem\u003eRes Sq\u003c/em\u003e (2024). https://doi.org/10.21203/rs.3.rs-4638168/v1\u003c/li\u003e\n\u003cli\u003eCoonahan, E.\u003cem\u003e et al.\u003c/em\u003e Whole-genome surveillance identifies markers of Plasmodium falciparum drug resistance and novel genomic regions under selection in Mozambique. \u003cem\u003emBio\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, e0176823 (2023). https://doi.org/10.1128/mbio.01768-23\u003c/li\u003e\n\u003cli\u003eBolger, A. M., Lohse, M. \u0026amp; Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 2114-2120 (2014). https://doi.org/10.1093/bioinformatics/btu170\u003c/li\u003e\n\u003cli\u003eLi, H. \u0026amp; Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 1754-1760 (2009). https://doi.org/10.1093/bioinformatics/btp324\u003c/li\u003e\n\u003cli\u003eLi, H.\u003cem\u003e et al.\u003c/em\u003e The Sequence Alignment/Map format and SAMtools. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 2078-2079 (2009). https://doi.org/10.1093/bioinformatics/btp352\u003c/li\u003e\n\u003cli\u003eMcKenna, A.\u003cem\u003e et al.\u003c/em\u003e The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. \u003cem\u003eGenome Res\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 1297-1303 (2010). https://doi.org/10.1101/gr.107524.110\u003c/li\u003e\n\u003cli\u003ePoplin, R.\u003cem\u003e et al.\u003c/em\u003e Scaling accurate genetic variant discovery to tens of thousands of samples. \u003cem\u003ebioRxiv\u003c/em\u003e, 201178 (2018). https://doi.org/10.1101/201178\u003c/li\u003e\n\u003cli\u003ePurcell, S.\u003cem\u003e et al.\u003c/em\u003e PLINK: a tool set for whole-genome association and population-based linkage analyses. \u003cem\u003eAm J Hum Genet\u003c/em\u003e \u003cstrong\u003e81\u003c/strong\u003e, 559-575 (2007). https://doi.org/10.1086/519795\u003c/li\u003e\n\u003cli\u003eAlexander, D. H., Novembre, J. \u0026amp; Lange, K. Fast model-based estimation of ancestry in unrelated individuals. \u003cem\u003eGenome Res\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 1655-1664 (2009). https://doi.org/10.1101/gr.094052.109\u003c/li\u003e\n\u003cli\u003eSchaffner, S. F., Taylor, A. R., Wong, W., Wirth, D. F. \u0026amp; Neafsey, D. E. hmmIBD: software to infer pairwise identity by descent between haploid genotypes. \u003cem\u003eMalaria Journal\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 196 (2018). https://doi.org/10.1186/s12936-018-2349-7\u003c/li\u003e\n\u003cli\u003eHenden, L., Lee, S., Mueller, I., Barry, A. \u0026amp; Bahlo, M. Identity-by-descent analyses for measuring population dynamics and selection in recombining pathogens. \u003cem\u003ePLOS Genetics\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, e1007279 (2018). https://doi.org/10.1371/journal.pgen.1007279\u003c/li\u003e\n\u003cli\u003eGautier, M. \u0026amp; Vitalis, R. rehh: an R package to detect footprints of selection in genome-wide SNP data from haplotype structure. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 1176-1177 (2012). https://doi.org/10.1093/bioinformatics/bts115\u003c/li\u003e\n\u003cli\u003eMi, H. \u0026amp; Thomas, P. PANTHER pathway: an ontology-based pathway database coupled with data analysis tools. \u003cem\u003eMethods Mol Biol\u003c/em\u003e \u003cstrong\u003e563\u003c/strong\u003e, 123-140 (2009). https://doi.org/10.1007/978-1-60761-175-2_7\u003c/li\u003e\n\u003cli\u003eFrichot, E. \u0026amp; Fran\u0026ccedil;ois, O. LEA: An R package for landscape and ecological association studies. \u003cem\u003eMethods in Ecology and Evolution\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 925-929 (2015). https://doi.org/https://doi.org/10.1111/2041-210X.12382\u003c/li\u003e\n\u003cli\u003eSzklarczyk, D.\u003cem\u003e et al.\u003c/em\u003e STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. \u003cem\u003eNucleic Acids Res\u003c/em\u003e \u003cstrong\u003e47\u003c/strong\u003e, D607-D613 (2019). https://doi.org/10.1093/nar/gky1131\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1. Regional Summary of\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003ePlasmodium falciparum\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;Genetic Diversity and Within-Host Complexity (N=17,565)\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"613\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 152px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRegion\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eN\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo. countries\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 56px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eF\u003csub\u003eWS\u003c/sub\u003e Median (Range)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 142px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMedian SNP Diversity (\u0026pi;)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 152px;\"\u003e\n \u003cp\u003eWest Africa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e5857\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 56px;\"\u003e\n \u003cp\u003e33.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e0.95 (0.23-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 142px;\"\u003e\n \u003cp\u003e1.62 x 10\u003csup\u003e-3\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 152px;\"\u003e\n \u003cp\u003eSoutheast Asia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e5790\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 56px;\"\u003e\n \u003cp\u003e32.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e0.98 (0.43-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 142px;\"\u003e\n \u003cp\u003e1.17 x 10\u003csup\u003e-3\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 152px;\"\u003e\n \u003cp\u003eEast Africa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e1604\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 56px;\"\u003e\n \u003cp\u003e9.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e0.95 (0.25-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 142px;\"\u003e\n \u003cp\u003e1.66 x 10\u003csup\u003e-3\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 152px;\"\u003e\n \u003cp\u003eSouth America\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e1562\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 56px;\"\u003e\n \u003cp\u003e8.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e0.99 (0.60-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 142px;\"\u003e\n \u003cp\u003e8.24 x 10\u003csup\u003e-4\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 152px;\"\u003e\n \u003cp\u003eSouth Asia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e1554\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 56px;\"\u003e\n \u003cp\u003e8.84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e0.97 (0.42-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 142px;\"\u003e\n \u003cp\u003e1.51 x 10\u003csup\u003e-3\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 152px;\"\u003e\n \u003cp\u003eCentral Africa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e514\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 56px;\"\u003e\n \u003cp\u003e2.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e0.90 (0.31-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 142px;\"\u003e\n \u003cp\u003e1.61 x 10\u003csup\u003e-3\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 152px;\"\u003e\n \u003cp\u003eOceania\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e343\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 56px;\"\u003e\n \u003cp\u003e1.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e0.97 (0.48-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 142px;\"\u003e\n \u003cp\u003e1.25 x 10\u003csup\u003e-3\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 152px;\"\u003e\n \u003cp\u003eSouth Central Africa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e255\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 56px;\"\u003e\n \u003cp\u003e1.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e0.77 (0.32-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 142px;\"\u003e\n \u003cp\u003e1.54 x 10\u003csup\u003e-3\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 152px;\"\u003e\n \u003cp\u003eHorn of Africa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 56px;\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e0.99 (0.61-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 142px;\"\u003e\n \u003cp\u003e1.51 x 10\u003csup\u003e-3\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7160640/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7160640/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"To investigate the global evolution and adaptation of Plasmodium falciparum, we analysed 17,565 isolates collected over three decades from 39 countries. This large-scale genomic study integrates identity-by-descent (IBD) networks, population structure and multi-layered association analyses to explore patterns of selection, transmission and drug resistance. Strong selection signals were observed at known resistance loci (e.g. pfcrt, pfdhps) and in genes linked to immune evasion and drug efflux (e.g. pfABCK1, pfMC-2TM), with signals varying by geography and time. In Southeast Asia, clonal expansion of the pfkelch13-C580Y mutation occurred and multi-layered analysis revealed co-occurrence with mutations in pfarps10, pfrad5, and pfMyoF, supporting a polygenic model of artemisinin resistance. In South America and Horn of Africa, elevated IBD was found in pfKIC7 and pfKIC9, interactors of pfkelch13, suggesting convergent evolution under drug pressure. New data from Brazil and Vietnam enhance resolution of global parasite diversity, highlighting the role of genomic surveillance in malaria control.","manuscriptTitle":"Global-scale population genetic analysis of Plasmodium falciparum identifies country- and region-specific patterns of malaria parasite adaptation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-18 08:35:49","doi":"10.21203/rs.3.rs-7160640/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"ce5a2934-7374-4dea-8ab4-9497349586de","owner":[],"postedDate":"August 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":52517100,"name":"Biological sciences/Genetics/Population genetics"},{"id":52517101,"name":"Biological sciences/Microbiology/Parasitology/Parasite genomics"}],"tags":[],"updatedAt":"2025-08-18T08:35:50+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-18 08:35:49","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7160640","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7160640","identity":"rs-7160640","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}