Impact of intensive control on malaria population genomics in Eastern Myanmar

Impact of intensive control on malaria population genomics in Eastern Myanmar | 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 Impact of intensive control on malaria population genomics in Eastern Myanmar Xue Li, Grace Arya, Aung Thu, Jordi Landier, Daniel Parker, Gilles Delmas, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6875020/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 13 Apr, 2026 Read the published version in Nature Microbiology → Version 1 posted You are reading this latest preprint version Abstract The malaria elimination program in Kayin State (Myanmar) utilizes malaria posts for rapid detection and treatment together with mass drug administration (MDA) in high transmission villages, and has reduced transmission by 97%. We examined the impact of control on parasite genomic parameters, using 2270 genome sequenced Plasmodium falciparum infections from 283 malaria posts, sampled over 58-months (2015 - 2020). Parasites were genetically depauperate: 1726 single-genotype infections comprised 166 unique genomes (≥90% IBD), while nine families (≥45% IBD) accounted for 62.5% of parasites sampled. Parasite effective population size decreased over the study period, but there was minimal change in artemisinin resistance alleles until 2020, when just one predominant genotype (carrying kelch13-R561H) remained. We observed sustained localized transmission of unique parasite genotypes revealing transmission chains: this resulted in positive correlations in parasite relatedness for ≤20 km. MDA resulted in parasite founder effects, providing genomic evidence for the efficacy of this control tool. These results reveal changes in population structure driven by control, and rapid shifts in allele frequency in a parasite population close to elimination. Biological sciences/Microbiology/Parasitology/Parasite genomics Health sciences/Diseases/Infectious diseases/Malaria Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Regions of low malaria transmission intensity predominate in Southeast (SE) Asia and South America and are becoming increasingly common in Africa 1 . A central challenge for malaria control is to develop efficient approaches to eliminate malaria from such regions. Rapid selection of drug-resistant parasites is a central concern for intensive malaria control programs. For example, kelch13 mutations conferring artemisinin resistance increased from 0-90% frequency in parasites collected from patients visiting Shoklo Malaria Research Unit (SMRU) clinics on the Thailand Myanmar border between 2003 – 2014 2 . Use of mass drug administration (MDA) is controversial for malaria treatment due to concerns about resistance. Prior use of chloroquine treated salt is thought to have accelerated selection of chloroquine resistance in the last century 3 . However, MDA is effective for treating submicroscopic malaria infections which comprise the majority of infections in many SE Asian locations 4 , and are missed by passive malaria surveillance. Submicroscopic infections may be cured effectively by MDA because there are few parasites per patient, and so treatment is more likely to be completely successful than in high-parasitemia infections. It has therefore been argued that MDA is not likely to promote resistance spread in low prevalence regions like SE Asia 5 . A previous paper examined the epidemiology of kelch13 haplotypes in Kayin State and showed limited changes of kelch13 artemisinin resistance alleles between 2013-2019 6 . Spread of resistance was much less rapid than occurred in the SMRU clinics (2000-2014); this is consistent with combined use of near-exhaustive coverage of communities with malaria posts and MDA imposing limited selection for ART-resistance. Here, we examine genomic epidemiology of P. falciparum samples collected between November 2015 and August 2020 during malaria elimination efforts in Kayin State. Between 2014-2020, the Malaria Elimination Task Force (METF) targeted four townships (Myawaddy, Kawkareik, Hlaingbwe, and Hpapun) in Kayin State, eastern Myanmar, for malaria elimination 7,8 . This was done using a combination of interventions: (i) 1475 village malaria posts (MPs) were opened, providing rapid diagnosis and malaria treatment (artemether–lumefantrine plus single low-dose primaquine); (ii) Mass Drug administration (MDA) (dihydroartemisinin–piperaquine [DHA-piperaquine] plus single dose primaquine once per month for 3 consecutive months) was used for 69 “hotspot” villages where malaria remained prevalent (>40% malaria and >20% Plasmodium falciparum ) 8 . The combination of these two approaches reduced P. falciparum cases by 97% from an incidence of 39 cases per 1000 person-years (May 2014–April 2015) 8 to 1 case per 1000 person-years (May 2019 to April 2020) 9 . Our central goal was to use genome sequence data to understand parasite transmission, population genomics and resistance evolution, and to use these data to inform future control efforts in elimination settings of SE Asia. Our key questions were to: (i) determine the distribution and stability of malaria populations within the Kayin State target region; (ii) to evaluate evidence for long-distance gene flow between Kayin State, Myanmar and other eastern SE Asian countries (Cambodia, Vietnam, Laos); (iii) document the origins of kelch13 resistance alleles; (iv) to determine the impact of MDA on parasite population structure; and (v) to evaluate appropriate genomic metrics for assessing transmission intensity. Results Extreme clonal expansion and inbreeding in a region under massive drug selection A total of 5014 dried blood spot (DBS) samples with geographic references from 413 MPs were collected between November 2015 and August 2020 as part of the METF malaria elimination program in eastern Myanmar, led by the SMRU. (Figure 1, Supplementary Figures 1 & 2). We processed and analyzed genome-wide sequencing data of 2270 DBS samples collected from 283 MPs (Table S1, Figure 1). After filtration of low-coverage samples and low-quality genotypes, the final Kayin State dataset contains 1927 P. falciparum samples with a set of 25,461 high-quality SNPs. 89.6% (1726/1927) of the samples were from single-genotype infections ( F ws ≥0.90, Supplementary Figure 4). To identify highly related individuals, we estimated pair-wise genetic relatedness ( r , proportion of genome that was IBD) and used these to cluster samples with ≥90% of the genome IBD ( r ≥0.90) (Figure 2). From the 1726 single-genotype Kayin samples, we identified 93 IBD clusters with unique genomes (2 to 229 samples per cluster), and 73 singletons, giving a total of 166 unique genomes. The Kayin population has an R G of 0.10 (166 unique genomes from 1726 single-genotype infected samples), containing 14 large clonal expansion clusters (>30 samples per cluster, Figure 2). In contrast, the R G ratios are much higher for the other SE Asian populations, with 0.72 (589/814) for SMRU clinics, 0.67 (123/184) for other Myanmar regions (not include Kayin), 0.54 (398/731) for Cambodia, 0.59 (102/173) for Viet Nam and 0.89 (80/90) for Laos. R G ratios range from 0.94 to 0.98 for African parasite populations (Figure 2B). Samples from the most common 10 IBD clusters account for 51.62% of the population, showing low parasite diversity in Kayin State. 151 out of 166 (91.0%) unique genomes were genetically related ( r ≥0.25) to at least one other unique genome, indicating high levels of inbreeding in the Kayin State parasite population (Figure 3A). We identified 9 closely related families, that account for 110/166 unique genomes. Each individual in a family has a r ≥0.45 with at least another family member. Samples from these families account for 65.2% (1126/1726) of all the single-genotype infections in Kayin State. To further estimate genealogical relationships of individuals inside each family, we analyzed the distribution of chromosomal IBD segments between closely related family members (Figure 3B, Supplementary Figure 5). For two of the families (family 1 and family 7), we were able to identify both parents and F1 progeny based on their chromosomal recombination patterns, indicating extremely small parasite population size in Kayin State. Localized transmission and regional stability of haplotypes Genetic relatedness in space. Spatial groups of IBD clusters reflect direct or indirect transmission chains of malaria parasite clones, so are particularly informative for understanding transmission dynamics. The IBD clusters and closely related families show localized spatial distribution (Figure 4A, Supplementary Figure 6). For example, 86% samples from the largest IBD cluster (carrying kelch13 R561H) were collected to the north of Hpapun Township. The second largest IBD cluster (carrying kelch13 F446I) was found mainly in the center of Hpapun Township, while the third largest IBD cluster (carrying kelch13 P441L) was found in the west of the same township. Similarly, different IBD clusters carrying kelch13 wildtype alleles, show localized distribution (Figure 4A). Spatial correlograms confirm that parasite relatedness is positively correlated at distances ≤20 km (Figure 4B). We also observed significant negative correlations in relatedness between 27.5-90km. The negative correlations further indicate the spatial pattern of parasite relatedness. Not only are parasites that are geographically proximate more likely to be related – those that are geographically distal (27.5-90km) are less likely to be related. Genetic relatedness in time. The length of time in which clonal lineages are sampled provides an indication of the frequency of outbreeding within malaria parasite populations. We detected clonal IBD clusters ( r ≥ 0.90, contain ≥2 samples) that were sampled across the 56-month study period, as well as new genome haplotypes generated through recombination (Supplementary Figure 7). Out of the 93 IBD clusters, 9 lasted ≥ 36 months (3 years of sampling), while 34 lasted ≤ 6 months. The mean duration of IBD clusters in Kayin population was 13.8 (1 se = 1.4) months. Furthermore, the mean sampling duration of closely related parasite family members ( r ≥0.45) was 48.7 (1 se = 3.0) months (Supplementary Figure 7), consistent with a low frequency of outbreeding in the population. We further analyzed the two parasite families for which both parents and progeny were identified. There were only 12 recombination events among the 109 samples (14 unique genomes, 41 months of family duration) from family 1, and 6 among the 31 samples (8 unique genomes, 48 months of family duration) from family 7 (Table S1, Supplementary Figure 5), consistent with low parasite outbreeding frequency. We used a temporal correlogram to examine the number of days over which correlations in relatedness were observed (Figure 4B). This revealed positive correlations in relatedness between parasites sampled ≤ 170 days apart. To evaluate the relative impact of space and time on parasite relatedness, we divided the control region into 50 regions using hierarchical clustering on principal components (HCPC) (Supplementary Figure 8); 29 of these HCPC regions contained from 10-161 parasites. We examined relatedness in parasites sampled within and between HCPC regions for parasites sampled parasites sampled 1-12, 13-24, 25-36 and 37-48 months apart (Figure 4C). We observed significantly greater relatedness among parasites sampled from the same HCPC unit, relative to those sampled from different HCPC units. This remained significant for parasites sampled up to ≤36 months apart, demonstrating spatial stability of parasite populations within Hpapun Township (Figure 4C). Long distance connectivity of parasite populations in SE Asia We used parasite relatedness to measure connectivity within west SE Asian populations and between west and east SE Asia (Supplementary Figure 9). We detected a high level of gene flow between parasite populations from SMRU clinics and Kayin State: 38.3% of Kayin samples had >25% IBD ( r > 0.25) with at least one sample from SMRU clinics, and 18.1% had >35% genome IBD ( r > 0.35). We identified two subpopulations, corresponding to west SE Asia (Kayin State, SMRU clinics and other Myanmar regions) and east SE Asia (Cambodia, Viet Nam and Laos) based on their genetic similarity and population structure (Supplementary Figure 10). Hence, we found no evidence for clonal transmission or recent recombination ( r > 0.15) between west and east SE Asia. The low connectivity between west and east SE Asia is further confirmed by the distribution of pfcrt mutations conferring piperaquine (PPQ) resistance. 52.19% of east SE Asian (Cambodia, Viet Nam and Laos) parasites carried PPQ-resistant pfcrt alleles between 2015 and 2018 (MalariaGEN Pf7 13 ). These included T93S (23.55%), I218F (11.87%) 14,15 , H97Y (5.37%), F145I (8.20%), G353V (2.125) 16 and G367C (1.08%) 17 . In contrast, these pfcrt mutations were absent from the Kayin dataset (Table S1) and from other west SE Asia regions (other Myanmar regions and west Thailand). Genomic measures of parasite population size We evaluated three genetic metrics (proportion of multiple-genotype infections, R G , and Ne ) in both Kayin State and SMRU parasite populations, for assessing how control efforts impact parasite population size (Figure 5). The malaria incidence decreased significantly in both regions over studying time. The incidence decreased from 273.9 cases per person-year in 2001 to 22.4 in 2011 for SMRU clinics 18 , and from 58.8 in 2016 to 1.0 in 2020 for Hpapun Township in the northern part of Kayin State, where >96% sequenced samples were collected 8,9 (Table S3). Proportion of multi-infections. The proportion of multiple clone infections (10.4%, 201/1927) in Kayin samples was low compared to other P. falciparum populations (Table S3) and remained low throughout the year (range: 3.7-15.1%). The proportion of multiple clone infections decreased significantly in SMRU clinics, from 34.3% in 2001 to 4.3% in 2014 ( p -value = 6.76e-05, R 2 = 0.78), consistent with a prior analysis 18 . However, this statistic showed no significant decline in Kayin ( p -value = 0.61, R 2 = 0.10, Figure 5B). R G . R G ratio is expected to be negatively correlated with the level of clonal expansion and positively correlated with transmission intensity 18 . R G ratios were lower in Kayin (0.10 to 0.30) between 2016 and 2020 than in SMRU clinics (range: 0.50-1.00) between 2001-2014. The R G ratio decreased over time in SMRU clinics ( p -value = 1.85e-03, R 2 = 0.78), but not in Kayin ( p -value = 0.54, R 2 = 0.14). Ne . We computed single-sample estimates of effective population size ( Ne ) using unique genomes from each population for each sampling year. Ne estimates were lower in Kayin (11.5 to 26.6) compared to SMRU clinics (15.5 to infinite). We detected significant reductions in Ne in both Kayin ( p -value = 3.69e-03, R 2 = 0.96) and SMRU clinics ( p -value = 0.03, R 2 = 0.40). The evolution of pfkelch13 alleles Impact of malaria elimination efforts on drug resistance. 61.32% of samples from Kayin State carried nonsynonymous SNP mutations in kelch13 (Supplementary Figure 11). The major mutant alleles were P441L (15.19%), F446I (15.01%), R561H (14.02%), and G449A (7.87%). Only 2.28% of Kayin samples carry C580Y. In comparison, in the adjacent SMRU clinics, C580Y was the dominant kelch13 mutant allele, reaching 71.05% allele frequency in 2014 (Table S2). There were 47 IBD clusters ( r ≥ 0.90) carrying mutant kelch13 alleles and 46 IBD clusters carrying wild-type kelch13 . We compared the size of IBD clusters carrying mutant kelch13 alleles with those carrying wild type kelch13 and found no significant difference (Figure 3C, Supplementary Figure 12). These results suggest that artemisinin selection was not the main driver for clonal expansion in Kayin State. Clonal expansion of parasite carrying kelch13 -R561H in 2020. Despite strong drug selection, frequencies of mutant kelch13 alleles remained stable between 2016 and 2019 6 (Figure 3C, Supplementary Figure 12). However, in 2020 one of the kelch13 alleles - R561H - reached 74.2%. This clonal expansion results from near elimination of parasites from most areas in Kayin, with the exception of northern Hpapun Township where parasites carrying kelch13 -R561H predominate (Figure 4D). 54.8% (40/73) of samples collected between January and August 2020 before the COVID-19 pandemic lockdowns were from one single malaria post (LH-0266B) (Table S1). The change in kelch13 allele frequencies were reflected by changes in diversity in this gene and its flanking regions. We measured expected heterozygosity ( He ) to quantify diversity. He in kelch13 remained high between 2016-2019 ( He = 0.78) but dropped to 0.41 in 2020. We observed parallel reductions in flanking region diversity with a drop from 0.48 to 0.19 between 2019 and 2020 (Supplementary Figure 11D). Origins of kelch13 alleles. We reconstructed the haplotypes surrounding the kelch13 gene (100kb upstream and 100 kb downstream). We found a wide variety of kelch13 genetic backgrounds, with one or more unique haplotypes per resistance allele (Supplementary Figure 13 & 14). Two P441L, one F446I, one G449A and one R561H haplotypes had shared ancestry between Kayin State and SMRU clinics or other Myanmar regions. However, none of these alleles had high frequency in SMRU clinics or other Myanmar regions. Two F446I and one C580Y haplotypes were uniquely observed in Kayin State, consistent with local origin. For two Kelch13 resistance alleles (G449A, R561H), the wildtype kelch13 haplotypes on which these resistance mutations arose were sampled in the early 2000s in SMRU clinics. Of the three C580Y haplotypes identified in Kayin, two were also found in SMRU clinics, while one was unique to Kayin State. The two C580Y haplotypes shared with SMRU clinics were only found to the south of Hpapun Township and 60-120km north of the SMRU clinics (Supplementary Figure 15). None of these haplotypes shared IBD with east SE Asia C580Y haplotypes (Supplementary Figure 16). Impact of mass drug usage on parasite population structure We predicted that MDA would reduce relatedness of pre and post MDA parasite populations due to clearance of the local parasite population and replacement with new parasite genotypes post-MDA, resulting in founder effects. There were three HCPC units in which sufficient parasites (n ≥ 20) were sequenced both pre and post MDA (Figure 6A). The relatedness between pre and post MDA parasites from these 3 HCPC units was significantly lower than observed between malaria parasites collected during the sampling time period from HCPC regions where MDA was not used (Figure 6B). Hence, MDA impacted parasite relatedness, consistent with efficient control of pre-MDA parasite genotypes, and post-MDA recolonization with unrelated genotypes. Discussion The METF elimination efforts, combining community malaria posts and MDA, significantly decreased malaria case numbers in the target area (near the Myanmar-Thailand border) between 2014-2020 8,9 . We described the key results from genomic surveillance during these elimination efforts, including parasite transmission patterns, population diversity and genomics, evolution of drug resistance, and genomic impacts of MDA using over 2000 whole genome sequenced P. falciparum samples collected between Nov 2015 – Aug 2020. Spatial and temporal structure of malaria populations Parasite sequences from the METF study region reveal extremely high levels of inbreeding and low levels of genetic variation. We found 166 genotype clusters (≥90% of the genome IBD) among the 1726 single clone samples sequenced. Hence only 10% of parasite genomes sampled are unique (R G = 0.1), and most infections show a clonal structure. In contrast, genetic richness is >0.94 in African populations sampled, and ranges from 0.54 to 0.89 in other SE Asian populations examined 19 . Furthermore, 110 of the 166 unique genomes are distributed among 9 different families (r ≥ 0.45). Parasites within these families typically carry one or two kelch13 alleles, that mostly likely inherited from the parents. This is clearly the case in the two families with both parents identified - family 1 (F446I and wild-type kelch13 ) and family 7 (C580Y and wild-type kelch13 ). Hence recombination is rare in these populations and most infections are clonally related. We observed strong spatial structure in the parasite population. This is particularly clear from the distribution of unique parasite genotypes. Such parasites result from self-fertilization of male and female gametes of the same genotype and allow spatial tracing of transmission networks. That these transmission networks are clustered in space is clear visually (Figure 4A), and statistically evident from autocorrelation analyses (Figure 4B), which reveal positive correlations in relatedness between genotypes for up to 20km. The local distribution of unique genotypes indicates (i) local transmission, and very few long-distance transmission events and (ii) reintroduction of circulating genotypes from asymptomatic carriers. Unique parasite genotypes were long lived in this low transmission setting, with some IBD clusters observed over the complete 56-month study period. This is clear from (i) the temporal autocorrelation, which reveals positive correlation in relatedness between parasites collected 170 days apart; (ii) from the elevated relatedness observed in parasites collected from the same HCPC regions but up 3 years apart. The strong spatial and temporal sub-structure of parasite populations in Kayin State is comparable to that observed in Cambodia 20 and Guyana 21 . A clonal expansion of parasites carrying kelch13 -R561H in 2020 As elimination approaches, genetic drift is expected to play an increasingly important role and expansions of parasite lineages may occur 22 . We observed a sudden increase in the frequency of IBD cluster 1 parasites carrying kelch13 -R561H in 2020. In this case, the increase of kelch13 -R561H frequency resulted from elimination of malaria from all regions of Kayin other than northern Hpapun Township (Figure 4A), where IBD cluster 1 carrying kelch13 -R561H is at high frequency. Despite predominating in northern Hpapun Township since 2017, the kelch13 -R561H didn’t spread into other areas of Kayin State. The rapid frequency increase of IBD cluster 1 in 2020 is consistent with bottlenecks and genetic drift in a P. falciparum population nearing elimination. Further sampling will determine whether this parasite genotype spreads further in Kayin State and elsewhere in Myanmar and Thailand. In contrast, Wasakul et al 22 describe a classic outbreak driven by a selective sweep in Laos, where a lineage carrying kelch13 -R539H (named LAA1) rose from a low frequency to replace the previously dominant KEL1/PLA1 (C580Y) population. MDA impacts parasite population structure Two control measures were used in Kayin State by METF: malaria posts (early diagnosis and community case management), and regional MDA in malaria hotspots 8 . The combination of these approaches significantly decreased malaria incidence 8,9 . Encouragingly, there was minimal evidence of selection for drug-resistance parasites through these elimination efforts which agrees with observations by Imwong et al 23 , McLean et al 24 and Thu et al 6 . At the genomic level, these combined control measures reduced parasite effective population size (Figure 5). This study provided an opportunity to examine the impact of one component of this elimination strategy - MDA - on parasite population structure. MDA is expected to generate bottlenecks between pre and post MDA malaria populations, because reservoirs of asymptomatic malaria are removed. We therefore expect to see founder effects resulting from newly colonizing parasites and large divergence between pre and post MDA populations, when compared to populations with no MDA. Our power to detect an impact was limited because most parasites sequenced were collected after MDA: only three HCPC units had sufficient numbers of infection sampled both pre and post MDA to examine this hypothesis. Nevertheless, as predicted, we saw lower relatedness between pre and post treatment parasites in these three HCPC regions than in control regions where MDA was not implemented. These results provide genomic evidence for the effectiveness of MDA in removing local parasite populations through effective clearance of both asymptomatic and symptomatic parasite infections. Genomic metrics for assessing transmission intensity Genetic metrics, such as proportion of multiple clone infections can provide useful metrics for assessing control efforts 18,25,26 . Such metrics are particularly useful in low transmission regions, where high prevalence of low-density asymptomatic infections complicates assessment of transmission using standard epidemiological methods. However, genetic metrics perform poorly when transmission levels are extremely low. In Senegal 25,26 , complexity of infection provided an unreliable metric for evaluating transmission when transmission is incidence < 100 cases per 1000 person-years. Consistent with this, we observed that both R G and proportion of Multiple infections worked poorly in Kayin where incidence ranged from 1 - 39 cases per 1000 person years. However, we found that another metric, effective population size (N e ) calculated from LD between unlinked markers, showed significant decline in both the Kayin State parasite population and from SMRU clinics. We suggest that N e may be a useful genomic indicator of transmission dynamics, particularly in parasite populations in which transmission has been reduced to near elimination levels. N e is typically used in conservation biology to assess viability in endangered populations of animals and plants. Our results suggest that this metric may also have utility for assessing whether parasite populations are approaching local extinction. No evidence for long-distance gene flow between Kayin State and Eastern SE Asian countries The current frontline treatments for P. falciparum malaria parasites have been failing in east SE Asia 23,27,28 , due to the spread of the multidrug-resistance parasites carrying kelch13 -C580Y mutation and plasmepsin 2 amplifications, named KEL1/PLA1. The KEL1/PLA1 lineage was first detected in Cambodia as the DHA-piperaquine was heavily used. When Cambodia withdraw DHA-piperaquine and adopted to artesunate–mefloquine, KEL1/PLA1 subgroups with acquired pfcrt mutations conferring piperaquine resistance rapidly spread to other ESEA countries, such as Laos and Vietnam 27 . There is a concern that this parasite lineage will further spread to west SE Asia, which has the majority of malaria cases in SE Asia 24 and where DHA-piperaquine is the frontline treatment for P. falciparum . Two lines of evidence suggest minimal geneflow between east and west SE Asia. First, we did not detect pfcrt mutations associated with piperaquine resistance on Thailand-Myanmar border or in the Kayin State sampling sites. Second, examination of genome-wide IBD sharing among 3718 infections (1458 unique genomes) revealed no recent recombination or clonal transmission between east and west SE Asia (Supplementary Figure 9). Origins of kelch13 resistance alleles kelch13 mutations conferring artemisinin resistance are established in both east and west SE Asia. C580Y is the major mutation in regions other than northern Myanmar, where F446I predominates 23 . In contrast, the dominant kelch13 mutations in the Kayin State include P441L, F446I, R561H, and G449A, depending on the location (Figure 4). While the majority of infections from nearby SMRU clinics carry C580Y (71.05% in 2014), the C580Y frequency in Kayin State was only 2.28%. The most widespread F446I haplotypes in Kayin State originated independently from the dominant F446I haplotype in northern Myanmar. What factors lead to the patterns of artemisinin resistance evolution seenin Kayin? Longitudinal studies in both Cambodia and from SMRU clinics have revealed that multiple independent kelch13 mutations emerged and spread initially (soft selective sweeps). Single kelch13 genotypes (typically kelch13 -C580Y) eventually outcompete other lineages leading to hard selective sweeps 2,20,23,27 . In contrast, we found limited evidence that strong drug selection drove drug resistance evolution in Kayin State: (i) we found no significant increase in kelch13 mutant allele frequencies before 2020 (Figure 3C, Supplementary Figure 11) 6 ; (ii) the size of clonal clusters was not significantly different when comparing kelch13 wildtype and mutant parasites (Figure 3C, Supplementary Figure 12). The small effective population size of malaria parasite populations may contribute to the patterns observed, because selection is inefficient when population sizes are small and genetic drift is enhanced 29 . The initial effective population size of malaria parasites in the Kayin State dataset was much smaller (Ne = 11.5 to 26.6) compared to SMRU clinics (15.5 – infinite) (Figure 2, Figure 5). Other factors that may also limit the impact of drug selection. Human population movements were more limited in Kayin compared to nearby SMRU clinics, especially in Northern Hpapun Township where human movement is limited by difficult terrain, the heavily militarized landscape, and a lack of year-round roads 7,30 , which can hinder transmission of resistance alleles. Similarly, low levels of recombination in Kayin State limits the rate of formation of new multi-locus parasite genotypes. The small parasite population size, limited population movement, and minimal recombination enhance the role of genetic drift rather than selection in determining drug resistance evolution in the Kayin State region. Our results, and those from other studies 23 illustrate how genetic drift can result in rapid stochastic changes in parasite population genomics and drug resistance status in small parasite populations close to elimination. This study had several limitations: (i) We analyzed malaria genomes collected from 2015 onwards. However, control efforts began earlier than this in 2014. Hence, we were unable to examine malaria population structure and diversity prior to initiation of control efforts. (ii) Use of finger prick blood samples and whole genome amplification resulted in bias towards sequencing high parasitemia infections. (iii) we were unable to score copy number variants, in genes such as Plasmepsin II/III, associated with piperaquine resistance from whole genome amplified DNA. However, the sustained sampling of a high proportion of blood spots collected over a 5-year period provides a unique dataset for examining impact of malaria control efforts on parasite population structure and resistance evolution. Methods Study area and sample origins The samples for this analysis were collected during routine diagnosis and treatment efforts in Kayin State as part of the METF malaria elimination effort led by the Shoklo Malaria Research Unit (SMRU, based on the Thailand-Myanmar border) (Figure 1, Supplementary Figure 1& 2). This METF project was established in 2014 and utilized two primary P. falciparum -focused interventions: the establishment of a large network of community-based malaria diagnosis and treatment posts (MPs), and targeted MDA in communities determined to have a high prevalence of asymptomatic P. falciparum infections. The MPs were stocked with filter papers (Whatman 3mm blotting paper) and were asked to collect dried blood spots (DBSs) from finger prick blood samples for patients with rapid diagnostic test (RDT) confirmed P. falciparum infection. Each DBS sample is linked to the MP from which it originated, and all MPs have geographic references (latitude and longitude). 5014 DBS samples were collected between November 2015 and August 2020 (Table S1). The DBS samples were then transported to SMRU and subsequently shipped to the Texas Biomedical Research Institute (in the U.S.A.) for molecular analyses. Sequencing library preparation We extracted DNA from the dried blood spots and enriched parasite genomes using selective whole genome amplification (sWGA) to following Li et al 31 and Oyola et al 32 . We extracted and purified genomic DNA using QIAamp® 96 DNA Blood Kit or QIAamp DNA Mini Kit, following the instruction manual for DNA isolation from dried blood spots. The DNA was then eluted in 100ul of 10mM Tris-HCl (pH 8.0-8.5) buffer. We used real-time quantitative PCR (qPCR) to estimate the numbers of parasite genomes in each DNA sample as described in Li et al 31 . For samples with more than 200 copies of parasite genome per ul, we used selective whole genome amplification (sWGA) to enrich parasite DNA. sWGA reactions were performed following Oyola et al 32 . Each 25 μl reaction contained at least 1000 copies of Plasmodium DNA, 1× BSA (New England Biolabs), 1 mM dNTPs (New England Biolabs), 3.5 μM of each amplification primer, 1× Phi29 reaction buffer (New England Biolabs), and 15 units of Phi29 polymerase (New England Biolabs). We used a PCR machine (SimpliAmp, Applied Biosystems) programmed to run a “stepdown” protocol: 35 °C for 10 min, 34 °C for 10 min, 33 °C for 10 min, 32 °C for 10 min, 31 °C for 10 min, 30 °C for 6 h then heating at 65 °C for 10 min to inactivate the enzymes prior to cooling to 4 °C. Sample were cleaned with AMPure XP Beads (Beckman Coulter), at a 1:1 ratio. We used the Quant-iT™ PicoGreen® Assay (Invitrogen) to determine the total amount of sWGA product, and quantified the proportion of Plasmodium DNA by qPCR. Only sWGA products with more than 50% DNA from Plasmodium were used for further library preparation and Illumina sequencing. We constructed PCR-free next generation sequencing libraries using 300ng sWGA product following the KAPA HyperPlus Kit protocol. All libraries were sequenced to an average coverage of 60× using Illumina Hiseq X or Novaseq sequencers. Whole-genome sequencing data generation We individually mapped whole-genome sequencing reads for each library against the P. falciparum 3D7 reference genome (PlasmoDB, release 46) and human GRCh38 reference genome, using the alignment algorithm BWA mem (http://bio-bwa.sourceforge.net/) under the default parameters. The resulting alignments were then converted to SAM format, sorted to BAM format, and deduplicated using picard tools v2.0.1 (http://broadinstitute.github.io/picard/). Reads mapping to the human genome were discarded before genotyping. We used Genome Analysis Toolkit GATK v3.7 (https://software.broadinstitute.org/gatk/) to recalibrate the base quality score based on a set of verified known variants 33 . We called variants for each sample using HaplotypeCaller, and calls from every 100 samples were merged using CombineGVCFs with default parameters. Variants were further called at all sample-level using GenotypeGVCFs , with parameters: --max_alternate_alleles 6 --variant_index_type LINEAR --variant_index_parameter 128000 --sample_ploidy 2 -nt 20. The recalibrated variant quality scores (VQSR) were calculated by comparing the raw variant distribution with the known and verified Plasmodium variant dataset. SNPs and indes with VQSR less than 1 or located outside of the core genome regions (21 Mb, defined in 33 ) were removed from further analysis. Samples with less than 50% of the core genome callable were also excluded from further analysis. Only biallelic SNPs that pass all the quality filter were used, unless otherwise specified. The final variants in VCF format were annotated at functional effect to genes and proteins using snpEff v4.3 (https://pcingola.github.io/SnpEff/) with 3D7 (PlasmoDB, release 46) as reference. We initially identified 1,302,006 single-nucleotide polymorphisms (SNPs) and 703,138 indels (Figure 1C). We removed 343 samples with > 20% genotypes missing. We then filtered the SNP calls following a “stringent” filtering method 34 , to generate a final list of 447,435 high-quality, biallelic, core-genome located (defined in 33 ) SNPs. To analyze complexity of infection and population structure, we further removed SNPs that were genotyped in less than 50% of samples or with minor allele frequency (MAF) 0.9 were assumed to come from single-genotype infections for samples from Kayin State. Allele frequencies across the genome were plotted and manually inspected to detect further possible complex infections. Relationships among parasite genotypes We used relatedness - r , defined as the fraction of the genome that is identical-by-descent (IBD) between a pair of individuals 36,37 - to estimate parasite relationships. Based on the distribution of relatedness among F1 progeny from malaria parasite genetic crosses (Supplementary Figure 3), we assume that parasites are genetically related if ≥ 25% of their genome is identical ( r ≥ 0.25); parasites are closely related (such as siblings or parent and progeny) if their relatedness is greater than 45% ( r ≥ 0.45). We considered samples to be clonal if their relatedness is over 90% ( r ≥ 0.90). We visualized relatedness among samples using the R package pheatmap and the Cytoscape software. We also examined the recombination patterns between closely related parasites and plotted shared IBD regions between estimated parents and progeny using karyoploteR . Surveillance of kelch13 haplotypes We extracted SNPs distributed within 100 kb upstream and 100 kb downstream of the kelch13 gene. We measured expected heterozygosity ( He ) at the kelch13 locus by treating kelch13 as a single locus with multiple alleles. We also measured He over the 200kb kelch13 haplotype region. To compare the relationships between different kelch13 haplotypes, we measured pairwise IBD sharing among all kelch13 haplotypes. We assume that haplotypes with IBD sharing ≥ 0.90 originated from the same mutational event; that when 0.35 ≤ IBD < 0.90, there was a one least recombination event to break the original haplotype; and when IBD < 0.35, these haplotypes have emerged independently. Comparisons of malaria parasite populations We compared the Kayin State parasite population with other world-wide malaria parasite populations (Figure 1, Table S2). The SMRU clinics are located around Mae Sot, in Tak Province along the international Thailand-Myanmar border. We used “other Myanmar” to represent sampling sites in Myanmar but not from Kayin State. West SE Asia population includes samples from Kayin State, SMRU clinics and other Myanmar regions, while east SE Asia population includes Cambodia, Viet Nam, and Laos. We merged raw SNP genotypes from the Kayin dataset with those from MalariaGEN P. falciparum Community Project 19 (release 6). We performed “stringent” filtration as described above, and selected loci with minor allele frequency > 0.05. We calculated genetic richness [R G = (G-1)/(S-1)] 38,39 to quantify the richness of clonal parasites in each population, where G is the number of unique genomes, and S is the total number of single genotype infected samples. For samples with relatedness > 0.9, only one representative sample per population with the highest genotype rate was selected and used for further analysis (Table S2). We pruned SNPs for linkage disequilibrium (LD) and generated a pairwise genetic distance matrix using PLINK with default parameters. We conducted principal component analyses (PCA) and ADMIXTURE analyses based on the pruned genotypes and distance matrix. We measured the proportion of pairs IBD across the genome within populations following the scripts in isoRelate 40 . We estimated effective population size ( Ne ) based on patterns of LD at unlinked loci, using methods implemented in NeEstimator v2.0 41 . Statistical analysis All statistical analysis was performed using R version 4.1.3. For pairwise comparisons between groups, we used Welch Two Sample T-test. We measured correlations between parasite genetic relatedness and geographic distance or time using the Mantel statistic using the mantel function in the vegan package. p <0.05 was considered statistically significant. We used hierarchical clustering on principal components (HCPC) following scripts in FactoMineR 42 to divide the 283 malaria posts with samples sequenced into 50 HCPC regions based on latitude and longitude. We then compared parasite relatedness within and between HCPC regions for parasites collected in the same year, between 1-2, 2-3 and 3-4 years apart. We compared relatedness between parasites collected from HCPC regions 6 months before and 6 months after MDA. As controls, we examined relatedness of parasites collected from HCPC regions where MDA was not used during the same time windows. Declarations Acknowledgments This work was supported by National Institutes of Health (NIH) grant R37 AI048071 (to TJCA) and P01 AI127338. Work at Texas Biomedical Research Institute was conducted in facilities constructed with support from Research Facilities Improvement Program grant C06 RR013556 from the National Center for Research Resources. SMRU is part of the Mahidol Oxford University Research Unit supported by the Wellcome Trust of Great Britain. The malaria elimination program (Malaria Elimination Task Force, METF) in Kayin State, Myanmar is supported by the Regional Artemisinin Initiative (Global Fund to Fight AIDS, Tuberculosis and Malaria) and the Bill and Melinda Gates Foundation (OPP1117507). The authors would like to acknowledge the contribution from all member of METF and SMRU, collaborators, and colleagues who have supported the elimination program. Author Contributions XL, FN, and TJCA conceived and designed the study. JL, AMT, GD, DMP, KML, KS and FN coordinated sample and data collection. XL, GAA, and AR processed samples, and generated genomic data. XL analyzed and interpreted the sequencing data, with input from TJCA, DMP, JL and FN. KML, KS, JL, DMP, and FN were involved in the management and coordination of the genetic surveillance project. XL and TJCA wrote the initial manuscript. XL, DMP, JL, FN and TJCA revised the manuscript. XL, DMP, FN and TJCA accessed and verified all the data. All authors provided critical revision of the manuscript. All authors had full access to all the data in the study and had final responsibility for the decision to submit for publication. TJCA and FN contributed equally. Competing Interests Authors declare that they have no competing interests. Additional Information Supplementary Information is available for this paper. Data Availability Raw sequencing data for the 2270 sequenced samples collected by the Malaria Elimination Task Force project from Myanmar used in the present analysis have been submitted to the NABI Sequence Read Archive (SRA, https://www.ncbi.nlm.nih.gov/sra) under the project number of PRJNA864839. The analysis code and data matrices (genetic distances, geographic distances and temporal distances) are available at: https://github.com/emilyli0325/Malaria-genomics-in-Eastern-Myanmar. This publication also uses data from the MalariaGEN P falciparum Community Project 19 . Funding National Institutes of Health grant R37 AI048071 (TJCA); National Institutes of Health grant 2P01AI127338 (TJCA: Core B PI); The malaria elimination program (Malaria Elimination Task Force, METF) in Kayin State, Myanmar is supported by the Regional Artemisinin Initiative (Global Fund to Fight AIDS, Tuberculosis and Malaria) and the Bill and Melinda Gates Foundation (OPP1117507) (FN); For the purpose of open access, the author has applied a CC BY public copyright license to any Author Accepted Manuscript version arising from this submission. References Venkatesan, P. The 2023 WHO World malaria report. Lancet Microbe 5 , e214, doi:10.1016/S2666-5247(24)00016-8 (2024). Anderson, T. J. et al. Population Parameters Underlying an Ongoing Soft Sweep in Southeast Asian Malaria Parasites. Mol Biol Evol 34 , 131-144, doi:10.1093/molbev/msw228 (2017). Wellems, T. E. & Plowe, C. V. Chloroquine-resistant malaria. J Infect Dis 184 , 770-776, doi:10.1086/322858 (2001). Imwong, M. et al. Numerical Distributions of Parasite Densities During Asymptomatic Malaria. J Infect Dis 213 , 1322-1329, doi:10.1093/infdis/jiv596 (2016). White, N. J. Does antimalarial mass drug administration increase or decrease the risk of resistance? Lancet Infect Dis 17 , e15-e20, doi:10.1016/S1473-3099(16)30269-9 (2017). Thu, A. M. et al. Molecular markers of artemisinin resistance during falciparum malaria elimination in Eastern Myanmar. Malar J 23 , 138, doi:10.1186/s12936-024-04955-6 (2024). Parker, D. M. et al. Scale up of a Plasmodium falciparum elimination program and surveillance system in Kayin State, Myanmar. Wellcome Open Res 2 , 98, doi:10.12688/wellcomeopenres.12741.2 (2017). Landier, J. et al. Effect of generalised access to early diagnosis and treatment and targeted mass drug administration on Plasmodium falciparum malaria in Eastern Myanmar: an observational study of a regional elimination programme. Lancet 391 , 1916-1926, doi:10.1016/S0140-6736(18)30792-X (2018). Legendre, E. et al. 'Forest malaria' in Myanmar? Tracking transmission landscapes in a diversity of environments. Parasit Vectors 16 , 324, doi:10.1186/s13071-023-05915-w (2023). Shomuyiwa, D. O. et al. Cabo Verde's malaria-free certification: A blueprint for eradicating malaria in Africa. J Taibah Univ Med Sci 19 , 534-536, doi:10.1016/j.jtumed.2024.04.001 (2024). Young, N. W. et al. High frequency of artemisinin partial resistance mutations in the great lake region revealed through rapid pooled deep sequencing. medRxiv , doi:10.1101/2024.04.29.24306442 (2024). Rosenthal, P. J. et al. The emergence of artemisinin partial resistance in Africa: how do we respond? Lancet Infect Dis , doi:10.1016/S1473-3099(24)00141-5 (2024). MalariaGen et al. Pf7: an open dataset of Plasmodium falciparum genome variation in 20,000 worldwide samples. Wellcome Open Res 8 , 22, doi:10.12688/wellcomeopenres.18681.1 (2023). Boonyalai, N. et al. Piperaquine resistant Cambodian Plasmodium falciparum clinical isolates: in vitro genotypic and phenotypic characterization. Malar J 19 , 269, doi:10.1186/s12936-020-03339-w (2020). Small-Saunders, J. L. et al. Evidence for the early emergence of piperaquine-resistant Plasmodium falciparum malaria and modeling strategies to mitigate resistance. PLoS Pathog 18 , e1010278, doi:10.1371/journal.ppat.1010278 (2022). Ross, L. S. et al. Emerging Southeast Asian PfCRT mutations confer Plasmodium falciparum resistance to the first-line antimalarial piperaquine. Nat Commun 9 , 3314, doi:10.1038/s41467-018-05652-0 (2018). Kane, J. et al. A Plasmodium falciparum genetic cross reveals the contributions of pfcrt and plasmepsin II/III to piperaquine drug resistance. bioRxiv , doi:10.1101/2023.06.06.543862 (2023). Nkhoma, S. C. et al. Population genetic correlates of declining transmission in a human pathogen. Mol Ecol 22 , 273-285, doi:10.1111/mec.12099 (2013). MalariaGen et al. An open dataset of Plasmodium falciparum genome variation in 7,000 worldwide samples. Wellcome Open Res 6 , 42, doi:10.12688/wellcomeopenres.16168.2 (2021). Miotto, O. et al. Multiple populations of artemisinin-resistant Plasmodium falciparum in Cambodia. Nature genetics 45 , 648-655, doi:10.1038/ng.2624 (2013). Vanhove, M. et al. Temporal and spatial dynamics of Plasmodium falciparum clonal lineages in Guyana. PLoS Pathog 20 , e1012013, doi:10.1371/journal.ppat.1012013 (2024). Wasakul, V. et al. Malaria outbreak in Laos driven by a selective sweep for Plasmodium falciparum kelch13 R539T mutants: a genetic epidemiology analysis. Lancet Infect Dis 23 , 568-577, doi:10.1016/S1473-3099(22)00697-1 (2023). Imwong, M. et al. Molecular epidemiology of resistance to antimalarial drugs in the Greater Mekong subregion: an observational study. Lancet Infect Dis 20 , 1470-1480, doi:10.1016/S1473-3099(20)30228-0 (2020). McLean, A. R. D. et al. Mass drug administration for the acceleration of malaria elimination in a region of Myanmar with artemisinin-resistant falciparum malaria: a cluster-randomised trial. Lancet Infect Dis 21 , 1579-1589, doi:10.1016/S1473-3099(20)30997-X (2021). Wong, W. et al. Evaluating the performance of Plasmodium falciparum genetic metrics for inferring National Malaria Control Programme reported incidence in Senegal. Malar J 23 , 68, doi:10.1186/s12936-024-04897-z (2024). Schaffner, S. F. et al. Malaria surveillance reveals parasite relatedness, signatures of selection, and correlates of transmission across Senegal. Nat Commun 14 , 7268, doi:10.1038/s41467-023-43087-4 (2023). Hamilton, W. L. et al. Evolution and expansion of multidrug-resistant malaria in southeast Asia: a genomic epidemiology study. Lancet Infect Dis 19 , 943-951, doi:10.1016/S1473-3099(19)30392-5 (2019). Amaratunga, C. et al. Dihydroartemisinin-piperaquine resistance in Plasmodium falciparum malaria in Cambodia: a multisite prospective cohort study. Lancet Infect Dis 16 , 357-365, doi:10.1016/S1473-3099(15)00487-9 (2016). Barton, N. H. Natural selection and random genetic drift as causes of evolution on islands. Philos Trans R Soc Lond B Biol Sci 351 , 785-794; discussion 795, doi:10.1098/rstb.1996.0073 (1996). Parker, D. M., Carrara, V. I., Pukrittayakamee, S., McGready, R. & Nosten, F. H. Malaria ecology along the Thailand-Myanmar border. Malar J 14 , 388, doi:10.1186/s12936-015-0921-y (2015). Li, X. et al. Genetic mapping of fitness determinants across the malaria parasite Plasmodium falciparum life cycle. PLoS Genetics 15 , e-1008453 (2019). Oyola, S. O. et al. Whole genome sequencing of Plasmodium falciparum from dried blood spots using selective whole genome amplification. Malaria journal 15 , 597 (2016). Miles, A. et al. Indels, structural variation, and recombination drive genomic diversity in Plasmodium falciparum. Genome research 26 , 1288-1299 (2016). McDew-White, M. et al. Mode and Tempo of Microsatellite Length Change in a Malaria Parasite Mutation Accumulation Experiment. Genome Biol Evol 11 , 1971-1985, doi:10.1093/gbe/evz140 (2019). Manske, M. et al. Analysis of Plasmodium falciparum diversity in natural infections by deep sequencing. Nature 487 , 375-379 (2012). Brown, T. S., Arogbokun, O., Buckee, C. O. & Chang, H. H. Distinguishing gene flow between malaria parasite populations. PLoS Genet 17 , e1009335, doi:10.1371/journal.pgen.1009335 (2021). 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 , 1-4 (2018). Eckert, C. G., Dorken, M. E. & Mitchell, S. A. Loss of Sex in Clonal Populations of a Flowering Plant, Decodon Verticillatus (Lythraceae). Evolution 53 , 1079-1092, doi:10.1111/j.1558-5646.1999.tb04523.x (1999). Echeverry, D. F. et al. Long term persistence of clonal malaria parasite Plasmodium falciparum lineages in the Colombian Pacific region. BMC Genet 14 , 2, doi:10.1186/1471-2156-14-2 (2013). Henden, L., Lee, S., Mueller, I., Barry, A. & Bahlo, M. Identity-by-descent analyses for measuring population dynamics and selection in recombining pathogens. PLoS Genet 14 , e1007279, doi:10.1371/journal.pgen.1007279 (2018). Do, C. et al. NeEstimator v2: re-implementation of software for the estimation of contemporary effective population size (Ne ) from genetic data. Molecular ecology resources 14 , 209-214, doi:10.1111/1755-0998.12157 (2014). Lê, S., Josse, J. & Husson, F. FactoMineR: an R package for multivariate analysis. Journal of statistical software 25 , 1-18, doi:10.18637/jss.v025.i01 (2008). Additional Declarations There is NO Competing Interest. Supplementary Files 2.Supplementarytables1306.11.2025.xlsx Supplementary Tables SupplementaryFigures.docx Cite Share Download PDF Status: Published Journal Publication published 13 Apr, 2026 Read the published version in Nature Microbiology → 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-6875020","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":476365808,"identity":"d2915748-0da1-41e3-a260-ccf4862f8159","order_by":0,"name":"Xue Li","email":"","orcid":"https://orcid.org/0000-0002-7501-4445","institution":"Disease Intervention and Prevention Program, Texas Biomedical Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Xue","middleName":"","lastName":"Li","suffix":""},{"id":476365809,"identity":"5a1fc774-3f28-45ae-9d02-a773ca316940","order_by":1,"name":"Grace Arya","email":"","orcid":"","institution":"Texas Biomedical Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Grace","middleName":"","lastName":"Arya","suffix":""},{"id":476365810,"identity":"04baf3f2-746b-4322-98d9-3d6fdb6fbe83","order_by":2,"name":"Aung Thu","email":"","orcid":"","institution":"Shoklo Malaria Research Unit","correspondingAuthor":false,"prefix":"","firstName":"Aung","middleName":"","lastName":"Thu","suffix":""},{"id":476365811,"identity":"466d1a8e-e2b9-45c4-9436-6fa25ff2f029","order_by":3,"name":"Jordi Landier","email":"","orcid":"","institution":"Aix Marseille Institute of Public Health","correspondingAuthor":false,"prefix":"","firstName":"Jordi","middleName":"","lastName":"Landier","suffix":""},{"id":476365812,"identity":"f5732883-482c-432f-bf84-ce30a9fc749c","order_by":4,"name":"Daniel Parker","email":"","orcid":"https://orcid.org/0000-0002-5352-7338","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Parker","suffix":""},{"id":476365813,"identity":"1d395942-a713-43ed-b532-c445bf28cb7b","order_by":5,"name":"Gilles Delmas","email":"","orcid":"","institution":"Shoklo Malaria Research Unit","correspondingAuthor":false,"prefix":"","firstName":"Gilles","middleName":"","lastName":"Delmas","suffix":""},{"id":476365814,"identity":"77b47c0d-21f6-433b-a7b5-0659554da252","order_by":6,"name":"Ann Reyes","email":"","orcid":"","institution":"Disease Intervention and Prevention Program, Texas Biomedical Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Ann","middleName":"","lastName":"Reyes","suffix":""},{"id":476365815,"identity":"183d0cf2-6245-4297-a89e-3effb39f4e07","order_by":7,"name":"Khin Lwin","email":"","orcid":"","institution":"Shoklo Malaria Research Unit","correspondingAuthor":false,"prefix":"","firstName":"Khin","middleName":"","lastName":"Lwin","suffix":""},{"id":476365816,"identity":"08fcdf5f-6108-45da-99ac-22098bb5cd6f","order_by":8,"name":"Kanlaya Sriprawat","email":"","orcid":"","institution":"Shoklo Malaria Research Unit","correspondingAuthor":false,"prefix":"","firstName":"Kanlaya","middleName":"","lastName":"Sriprawat","suffix":""},{"id":476365817,"identity":"4cb3d85f-c791-455d-ae80-cc0597d35dcd","order_by":9,"name":"François Nosten","email":"","orcid":"https://orcid.org/0000-0002-7951-0745","institution":"Shoklo Malaria Research Unit","correspondingAuthor":false,"prefix":"","firstName":"François","middleName":"","lastName":"Nosten","suffix":""},{"id":476365807,"identity":"88b8c8cb-0416-4912-977d-2ec9d4d00923","order_by":10,"name":"Timothy Anderson","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6UlEQVRIiWNgGAWjYJACCRDBz8DAxpAAYh0goJwHpkWyjWQtBseAWhiI0WLPfvbgjY976uSM7/eYPXhQsU2O7wDzsY9f8NnCk5dsOePZYWOzY2zpBglnbhtLHmBLni2D12E5ZtI8Bw4kbjvGfEwise124oYDPMbMEvi08L8xk/5zoC5xcxtjG0hLPWEtEkBbGA4wJ25gg9iSYADUwvgBn5Ybb4wtew4cNpY4lgb2i+HMw2zJzHh0MLD35xje+HGgTo6/+YzZwx8Vt+X5jjcfZvyBTw8mAFrBzEOaFiAg1ZZRMApGwSgY3gAALq1ONKRasB0AAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-0191-0204","institution":"Texas Biomedical Research Institute","correspondingAuthor":true,"prefix":"","firstName":"Timothy","middleName":"","lastName":"Anderson","suffix":""}],"badges":[],"createdAt":"2025-06-11 22:05:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6875020/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6875020/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41564-026-02327-1","type":"published","date":"2026-04-13T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":85448052,"identity":"fc03c1a1-86bb-4819-9c62-3f6400e3416d","added_by":"auto","created_at":"2025-06-26 03:32:12","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":663601,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSample collection and dataset summary.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Physical geography of the Malaria Elimination Task Force (METF) intervention region. The METF project was performed at four townships (Myawaddy, Kawkareik, Hlaingbwe, and Hpapun) of Kayin State (light yellow shaded region), Myanmar. Left panel, the distribution of malaria posts (MPs, green dots). Vermillion dots indicate locations where mass drug administration (MDAs) were applied. Middle panel, location of sequenced samples. Over 96% of the sequenced sample were from Hpapun township, the northern part of Kayin State. Right panel, elevation map. Elevation data was downloaded from the United States Geological Survey (USGS, https://earthexplorer.usgs.gov/). The locations of roads (brown) and rivers (light blue) were from Myanmar Information Management Unit (https://themimu.info/). (B) Temporal distribution of samples collected through the METF project. Red arrows indicate the time when MDAs were applied. (C) Analysis pipeline for samples collected from Kayin State. (D) World-wide malaria parasite datasets used in this study. Sequencing data other than those from Kayin State were from MalariaGEN (\u003ca href=\"https://www.malariagen.net/\"\u003ehttps://www.malariagen.net/\u003c/a\u003e, release 6). Shoklo Malaria Research Unit (SMRU) clinics are located at the Thailand Myanmar border. Samples from Myanmar but not from Kayin State are labeled as “other Myanmar”, see Figure S1 for detailed sample regions.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-6875020/v1/ae2be8af66927e2038cf10c6.png"},{"id":85448054,"identity":"7889ec47-478e-4a57-a9fe-d0e1d1659df3","added_by":"auto","created_at":"2025-06-26 03:32:12","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":811594,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eParasite relatedness and level of clonal transmission.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Heatmap showing relatedness among Kayin samples. Pairwise parasite relatedness (\u003cem\u003er\u003c/em\u003e) was measured as the proportion of genomes that are identical by decent (IBD) between pairs of samples. Samples with \u003cem\u003er\u003c/em\u003e≥ 0.9 are considered as IBD and share the same unique genome. Color bars at the top of the heatmap indicate information for each sample: MDA, if the sample came from a malaria post with (orange) or without (blue) mass drug administration; Year, the year of sampling; \u003cem\u003ekelch13\u003c/em\u003e, the genotype of\u003cem\u003e kelch13 \u003c/em\u003e- only alleles with frequency \u0026gt; 2% in at least one sampling year were colored. (B) Level of clonal transmission. Red numbers on top of bars indicate genetic richness. Kayin population had the highest level of clonal transmission compare to other populations. SMRU, Shoklo Malaria Research Unit; WSEA, west southeast Asia; ESEA, east southeast Asia.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-6875020/v1/2a45331c035dc0ac383ecaae.png"},{"id":85448055,"identity":"637837a8-5c78-440c-ae50-188a0a1bf976","added_by":"auto","created_at":"2025-06-26 03:32:12","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":294871,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eParasite clonal expansion and inbreeding in Kayin State.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) IBD network of unique genomes from Kayin population. Nodes: each circle indicates one unique genome and is color coded based on its \u003cem\u003ekelch13 \u003c/em\u003ealleles; circle size indicates sample size (ranged from 1 to 229). Edges: connections with relatedness (\u003cem\u003er\u003c/em\u003e) ≥ 0.25; thicker lines indicate higher relatedness; red lines are connections with \u003cem\u003er \u003c/em\u003e≥ 0.45. Parasites from closely related families (f1 to f9) are labeled using boxes. 152 of 166 unique genomes were include in the network, representing 98.6% of single-genotype infected samples. (B) Pedigree tree of parasites from family 1 (f1) and chromosome plot for an estimated progeny (MP3233). See Figure S7 for chromosome plots for all progeny. We infer that the parents of f1 are C13 (IBD cluster 13, \u003cem\u003ekelch13-\u003c/em\u003ewildtype) and C17 (IBD cluster 17, \u003cem\u003ekelch13\u003c/em\u003e-F446I). (C) Proportion of unique genomes across time. Each segment within a bar represents one unique genome, which is colored based on its \u003cem\u003ekelch13\u003c/em\u003e allele. Black blocks indicate number of unique genomes that were recovered only once (“singletons”). A clonal expansion of IBD cluster 1(\u003cem\u003ekelch13\u003c/em\u003e-R561H) parasites was detected in 2020.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-6875020/v1/2ce0444370e2b2fd174add46.png"},{"id":85448492,"identity":"b703dc4d-17c5-4565-9d7c-caed99dd31f8","added_by":"auto","created_at":"2025-06-26 03:40:12","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":435310,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLocalized transmission and temporal stability.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Spatial distribution of different IBD clusters. Each circle represents samples collected from an individual malaria post. IBD cluster IDs and corresponding \u003cem\u003ekelch13\u003c/em\u003e alleles are indicated at the bottom left of each panel; for example, 1_R561H indicates IBD cluster number 1 carrying the \u003cem\u003ekelch13\u003c/em\u003e R561H allele. (B) Correlogram analysis of pair-wise parasite relatedness across space and time. (C) Comparison of relatedness between within-group and between-group HCPC region pairs. X-axis indicates time intervals in month. (D) Temporal dynamics and spatial distribution of the IBD cluster 1 (\u003cem\u003ekelch13 \u003c/em\u003eR561H) through the sampling year.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-6875020/v1/659135b9a4fd7f5f62cc69f9.png"},{"id":85448057,"identity":"873bacfa-1f8f-43fe-93e0-e5c9f930cc23","added_by":"auto","created_at":"2025-06-26 03:32:12","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":239329,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGenomic measures of parasite population size.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Population size estimated using genetic richness (R\u003csub\u003eG\u003c/sub\u003e), proportion of samples from multiple-genotype infection (multi-infections%), and effective population size (Ne). (B) Comparison of genetic metrics between Kayin State and Thailand-Myanmar border (SMRU clinics) populations. The incidence data for Kayin State was from Landier et al., 2018 and Legendre et al., 2023; and the incidence data for SMRU clinics (regions near Mae Sot) was from Nkhoma et al., 2013. NS, not significant; SMRU, Shoklo Malaria Research Unit clinics.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-6875020/v1/a6e4af00803eb8c8a6601a0e.png"},{"id":85448053,"identity":"e9baf1eb-1e14-4860-83b6-1cc52bea7b25","added_by":"auto","created_at":"2025-06-26 03:32:12","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":76471,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe impact of mass drug administration (MDA).\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Population relatedness before and after MDA. MDAs were implemented between June 24, 2017 – October 12, 2017 in HCPC regions 30, 33 and 35, using DHA-piperaquine plus a single dose of primaquine administered monthly for three consecutive months. Bar segments represent unique genomes identified within the 6 months before (April 12, 2017 – October 12, 2017) or after the intervention (October 13, 2017 – April 12, 2018), color-coded by \u003cem\u003ekelch13\u003c/em\u003eallele type. (B) Comparison of parasite genetic relatedness between HCPC regions with and without MDA intervention. Parasites collected from the same time windows from HCPC regions where MDA was not used provided “no MDA” controls.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-6875020/v1/90ef799f5001cef8d39dcca1.png"},{"id":106854780,"identity":"ad913de7-8d2e-4125-ac65-78396f5273e9","added_by":"auto","created_at":"2026-04-14 07:13:28","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3414845,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6875020/v1/5ceb2c65-0030-469b-b787-b47a01a1bf18.pdf"},{"id":85448493,"identity":"219970b1-2e57-4189-b947-b853ff0e88df","added_by":"auto","created_at":"2025-06-26 03:40:12","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":552409,"visible":true,"origin":"","legend":"Supplementary Tables","description":"","filename":"2.Supplementarytables1306.11.2025.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6875020/v1/18bf7723639955cdd6f7fad0.xlsx"},{"id":85448059,"identity":"6f0d323c-dcee-4461-ab85-1b0dfe4224b3","added_by":"auto","created_at":"2025-06-26 03:32:12","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":3517245,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-6875020/v1/01e92f8cc931716ff22d28e9.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Impact of intensive control on malaria population genomics in Eastern Myanmar","fulltext":[{"header":"Introduction","content":"\u003cp\u003eRegions of low malaria transmission intensity predominate in Southeast (SE) Asia and South America and are becoming increasingly common in Africa \u003csup\u003e1\u003c/sup\u003e. A central challenge for malaria control is to develop efficient approaches to eliminate malaria from such regions. Rapid selection of drug-resistant parasites is a central concern for intensive malaria control programs. For example, \u003cem\u003ekelch13\u003c/em\u003e mutations conferring artemisinin resistance increased from 0-90% frequency in parasites collected from patients visiting Shoklo Malaria Research Unit (SMRU) clinics on the Thailand Myanmar border between 2003 \u0026ndash; 2014 \u003csup\u003e2\u003c/sup\u003e. \u0026nbsp;Use of mass drug administration (MDA) is controversial for malaria treatment due to concerns about resistance. Prior use of chloroquine treated salt is thought to have accelerated selection of chloroquine resistance in the last century \u003csup\u003e3\u003c/sup\u003e. However, MDA is effective for treating submicroscopic malaria infections which comprise the majority of infections in many SE Asian locations\u003csup\u003e4\u003c/sup\u003e, and are missed by passive malaria surveillance. Submicroscopic infections may be cured effectively by MDA because there are few parasites per patient, and so treatment is more likely to be completely successful than in high-parasitemia infections. It has therefore been argued that MDA is not likely to promote resistance spread in low prevalence regions like SE Asia \u003csup\u003e5\u003c/sup\u003e. A previous paper examined the epidemiology of \u003cem\u003ekelch13\u003c/em\u003e haplotypes in Kayin State and showed limited changes of \u003cem\u003ekelch13\u003c/em\u003e artemisinin resistance alleles between 2013-2019 \u003csup\u003e6\u003c/sup\u003e. Spread of resistance was much less rapid than occurred in the SMRU clinics (2000-2014); this is consistent with combined use of near-exhaustive coverage of communities with malaria posts and MDA imposing limited selection for ART-resistance.\u003c/p\u003e\n\u003cp\u003eHere, we examine genomic epidemiology of \u003cem\u003eP. falciparum\u003c/em\u003e samples collected between November 2015 and August 2020 during malaria elimination efforts in Kayin State. Between 2014-2020, the Malaria Elimination Task Force (METF) targeted four townships (Myawaddy, Kawkareik, Hlaingbwe, and Hpapun) in Kayin State, eastern Myanmar, for malaria elimination \u003csup\u003e7,8\u003c/sup\u003e. This was done using a combination of interventions: (i) 1475 village malaria posts (MPs) were opened, providing rapid diagnosis and malaria treatment (artemether\u0026ndash;lumefantrine plus single low-dose primaquine); (ii) Mass Drug administration (MDA) (dihydroartemisinin\u0026ndash;piperaquine [DHA-piperaquine] plus single dose primaquine once per month for 3 consecutive months) was used for 69 \u0026ldquo;hotspot\u0026rdquo; villages where malaria remained prevalent (\u0026gt;40% malaria and \u0026gt;20% \u003cem\u003ePlasmodium falciparum\u003c/em\u003e) \u003csup\u003e8\u003c/sup\u003e. The combination of these two approaches reduced \u003cem\u003eP. falciparum\u003c/em\u003e cases by 97% from an incidence of 39 cases per 1000 person-years (May 2014\u0026ndash;April 2015) \u003csup\u003e8\u003c/sup\u003e to 1 case per 1000 person-years (May 2019 to April 2020) \u003csup\u003e9\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur central goal was to use genome sequence data to understand parasite transmission, population genomics and resistance evolution, and to use these data to inform future control efforts in elimination settings of SE Asia. Our key questions were to: (i) determine the distribution and stability of malaria populations within the Kayin State target region; (ii) to evaluate evidence for long-distance gene flow between Kayin State, Myanmar and other eastern SE Asian countries (Cambodia, Vietnam, Laos); (iii) document the origins of \u003cem\u003ekelch13\u003c/em\u003e resistance alleles; (iv) to determine the impact of MDA on parasite population structure; and (v) to evaluate appropriate genomic metrics for assessing transmission intensity.\u0026nbsp;\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eExtreme clonal expansion and inbreeding in a region under massive drug selection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA total of 5014 dried blood spot (DBS) samples with geographic references from 413 MPs were collected between November 2015 and August 2020 as part of the METF malaria elimination program in eastern Myanmar, led by the SMRU. (Figure 1, Supplementary Figures 1 \u0026amp; 2). We processed and analyzed genome-wide sequencing data of 2270 DBS samples collected from 283 MPs (Table S1, Figure 1). After filtration of low-coverage samples and low-quality genotypes, the final Kayin State dataset contains 1927 \u003cem\u003eP. falciparum\u003c/em\u003e samples with a set of 25,461 high-quality SNPs. 89.6% (1726/1927) of the samples were from single-genotype infections (\u003cem\u003eF\u003csub\u003ews\u0026nbsp;\u003c/sub\u003e\u003c/em\u003e\u0026ge;0.90, Supplementary Figure 4).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo identify highly related individuals, we estimated pair-wise genetic relatedness (\u003cem\u003er\u003c/em\u003e, proportion of genome that was IBD) and used these to cluster samples with \u0026ge;90% of the genome IBD (\u003cem\u003er\u003c/em\u003e \u0026ge;0.90) (Figure 2). From the 1726 single-genotype Kayin samples, we identified 93 IBD clusters with unique genomes (2 to 229 samples per cluster), and 73 singletons, giving a total of 166 unique genomes. The Kayin population has an R\u003csub\u003eG\u003c/sub\u003e of 0.10 (166 unique genomes from 1726 single-genotype infected samples), containing 14 large clonal expansion clusters (\u0026gt;30 samples per cluster, Figure 2). In contrast, the R\u003csub\u003eG\u003c/sub\u003e ratios are much higher for the other SE Asian populations, with 0.72 (589/814) for SMRU clinics, 0.67 (123/184) for other Myanmar regions (not include Kayin), 0.54 (398/731) for Cambodia, 0.59 (102/173) for Viet Nam and 0.89 (80/90) for Laos. R\u003csub\u003eG\u0026nbsp;\u003c/sub\u003eratios range from 0.94 to 0.98 for African parasite populations (Figure 2B).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSamples from the most common 10 IBD clusters account for 51.62% of the population, showing low parasite diversity in Kayin State. 151 out of 166 (91.0%) unique genomes were genetically related (\u003cem\u003er\u003c/em\u003e \u0026ge;0.25) to at least one other unique genome, indicating high levels of inbreeding in the Kayin State parasite population (Figure 3A). We identified 9 closely related families, that account for 110/166 unique genomes. Each individual in a family has a \u003cem\u003er\u003c/em\u003e \u0026ge;0.45 with at least another family member. Samples from these families account for 65.2% (1126/1726) of all the single-genotype infections in Kayin State. To further estimate genealogical relationships of individuals inside each family, we analyzed the distribution of chromosomal IBD segments between closely related family members (Figure 3B, Supplementary Figure 5). For two of the families (family 1 and family 7), we were able to identify both parents and F1 progeny based on their chromosomal recombination patterns, indicating extremely small parasite population size in Kayin State.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLocalized transmission and regional stability of haplotypes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGenetic relatedness in space. Spatial groups of IBD clusters reflect direct or indirect transmission chains of malaria parasite clones, so are particularly informative for understanding transmission dynamics. The IBD clusters and closely related families show localized spatial distribution (Figure 4A, Supplementary Figure 6). For example, 86% samples from the largest IBD cluster (carrying \u003cem\u003ekelch13\u003c/em\u003e R561H) were collected to the north of Hpapun Township. The second largest IBD cluster (carrying \u003cem\u003ekelch13\u003c/em\u003e F446I) was found mainly in the center of Hpapun Township, while the third largest IBD cluster (carrying\u003cem\u003e\u0026nbsp;kelch13\u003c/em\u003e P441L) was found in the west of the same township. Similarly, different IBD clusters carrying \u003cem\u003ekelch13\u0026nbsp;\u003c/em\u003ewildtype alleles, show localized distribution (Figure 4A). Spatial correlograms confirm that parasite relatedness is positively correlated at distances \u0026le;20 km (Figure 4B). We also observed significant negative correlations in relatedness between 27.5-90km. The negative correlations further indicate the spatial pattern of parasite relatedness. Not only are parasites that are geographically proximate more likely to be related \u0026ndash; those that are geographically distal (27.5-90km) are less likely to be related.\u003c/p\u003e\n\u003cp\u003eGenetic relatedness in time. The length of time in which clonal lineages are sampled provides an indication of the frequency of outbreeding within malaria parasite populations. We detected clonal IBD clusters (\u003cem\u003er\u003c/em\u003e \u0026ge; 0.90, contain \u0026ge;2 samples) that were sampled across the 56-month study period, as well as new genome haplotypes generated through recombination (Supplementary Figure 7). Out of the 93 IBD clusters, 9 lasted \u0026ge; 36 months (3 years of sampling), while 34 lasted \u0026le; 6 months. The mean duration of IBD clusters in Kayin population was 13.8 (1 \u003cem\u003ese\u003c/em\u003e = 1.4) months. Furthermore, the mean sampling duration of closely related parasite family members (\u003cem\u003er\u003c/em\u003e \u0026ge;0.45) was 48.7 (1 \u003cem\u003ese\u003c/em\u003e = 3.0) months (Supplementary Figure 7), consistent with a low frequency of outbreeding in the population. We further analyzed the two parasite families for which both parents and progeny were identified. There were only 12 recombination events among the 109 samples (14 unique genomes, 41 months of family duration) from family 1, and 6 among the 31 samples (8 unique genomes, 48 months of family duration) from family 7 (Table S1, Supplementary Figure 5), consistent with low parasite outbreeding frequency.\u003c/p\u003e\n\u003cp\u003eWe used a temporal correlogram to examine the number of days over which correlations in relatedness were observed (Figure 4B). This revealed positive correlations in relatedness between parasites sampled \u0026le; 170 days apart. To evaluate the relative impact of space and time on parasite relatedness, we divided the control region into 50 regions using hierarchical clustering on principal components (HCPC) (Supplementary Figure 8); 29 of these HCPC regions contained from 10-161 parasites. We examined relatedness in parasites sampled within and between HCPC regions for parasites sampled parasites sampled 1-12, 13-24, 25-36 and 37-48 months apart (Figure 4C). We observed significantly greater relatedness among parasites sampled from the same HCPC unit, relative to those sampled from different HCPC units. This remained significant for parasites sampled up to \u0026le;36 months apart, demonstrating spatial stability of parasite populations within Hpapun Township (Figure 4C).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLong distance connectivity of parasite populations in SE Asia\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe used parasite relatedness to measure connectivity within west SE Asian populations and between west and east SE Asia (Supplementary Figure 9). We detected a high level of gene flow between parasite populations from SMRU clinics and Kayin State: 38.3% of Kayin samples had \u0026gt;25% IBD (\u003cem\u003er\u003c/em\u003e \u0026gt; 0.25) with at least one sample from SMRU clinics, and 18.1% had \u0026gt;35% genome IBD (\u003cem\u003er\u003c/em\u003e \u0026gt; 0.35). We identified two subpopulations, corresponding to west SE Asia (Kayin State, SMRU clinics and other Myanmar regions) and east SE Asia (Cambodia, Viet Nam and Laos) based on their genetic similarity and population structure (Supplementary Figure 10). Hence, we found no evidence for clonal transmission or recent recombination (\u003cem\u003er\u003c/em\u003e \u0026gt; 0.15) between west and east SE Asia.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe low connectivity between west and east SE Asia is further confirmed by the distribution of \u003cem\u003epfcrt\u003c/em\u003e mutations conferring piperaquine (PPQ) resistance. 52.19% of east SE Asian (Cambodia, Viet Nam and Laos) parasites carried PPQ-resistant\u003cem\u003e\u0026nbsp;pfcrt\u0026nbsp;\u003c/em\u003ealleles between 2015 and 2018 (MalariaGEN Pf7\u0026nbsp;\u003csup\u003e13\u003c/sup\u003e). These included T93S (23.55%), I218F (11.87%)\u003csup\u003e14,15\u003c/sup\u003e, H97Y (5.37%), F145I (8.20%), G353V (2.125)\u003csup\u003e16\u003c/sup\u003e and G367C (1.08%)\u003csup\u003e17\u003c/sup\u003e. In contrast, these \u003cem\u003epfcrt\u0026nbsp;\u003c/em\u003emutations were absent from the Kayin dataset (Table S1) and from other west SE Asia regions (other Myanmar regions and west Thailand).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGenomic measures of parasite population size\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe evaluated three genetic metrics (proportion of multiple-genotype infections, R\u003csub\u003eG\u003c/sub\u003e, and \u003cem\u003eNe\u003c/em\u003e) in both Kayin State and SMRU parasite populations, for assessing how control efforts impact parasite population size (Figure 5). The malaria incidence decreased significantly in both regions over studying time. The incidence decreased from 273.9 cases per person-year in 2001 to 22.4 in 2011 for SMRU clinics \u003csup\u003e18\u003c/sup\u003e, and from 58.8 in 2016 to 1.0 in 2020 for Hpapun Township in the northern part of Kayin State, where \u0026gt;96% sequenced samples were collected \u003csup\u003e8,9\u003c/sup\u003e (Table S3).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eProportion of multi-infections. The proportion of multiple clone infections (10.4%, 201/1927) in Kayin samples was low compared to other \u003cem\u003eP. falciparum\u003c/em\u003e populations (Table S3) and remained low throughout the year (range: 3.7-15.1%). The proportion of multiple clone infections decreased significantly in SMRU clinics, from 34.3% in 2001 to 4.3% in 2014 (\u003cem\u003ep\u003c/em\u003e-value = 6.76e-05, \u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e = 0.78), consistent with a prior analysis \u003csup\u003e18\u003c/sup\u003e. However, this statistic showed no significant decline in Kayin (\u003cem\u003ep\u003c/em\u003e-value = 0.61, \u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e = 0.10, Figure 5B).\u003c/p\u003e\n\u003cp\u003eR\u003csub\u003eG\u003c/sub\u003e. R\u003csub\u003eG\u003c/sub\u003e ratio is expected to be negatively correlated with the level of clonal expansion and positively correlated with transmission intensity \u003csup\u003e18\u003c/sup\u003e. R\u003csub\u003eG\u003c/sub\u003e ratios were lower in Kayin (0.10 to 0.30) between 2016 and 2020 than in SMRU clinics (range: 0.50-1.00) between 2001-2014. The R\u003csub\u003eG\u003c/sub\u003e ratio decreased over time in SMRU clinics (\u003cem\u003ep\u003c/em\u003e-value = 1.85e-03, \u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e = 0.78), but not in Kayin (\u003cem\u003ep\u003c/em\u003e-value = 0.54, \u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e = 0.14).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNe\u003c/em\u003e. We computed single-sample estimates of effective population size (\u003cem\u003eNe\u003c/em\u003e) using unique genomes from each population for each sampling year. \u003cem\u003eNe\u003c/em\u003e estimates were lower in Kayin (11.5 to 26.6) compared to SMRU clinics (15.5 to infinite). We detected significant reductions in \u003cem\u003eNe\u003c/em\u003e in both Kayin (\u003cem\u003ep\u003c/em\u003e-value = 3.69e-03, \u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e = 0.96) and SMRU clinics (\u003cem\u003ep\u003c/em\u003e-value = 0.03, \u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e = 0.40).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe evolution of \u003cem\u003epfkelch13\u003c/em\u003e alleles\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eImpact of malaria elimination efforts on drug resistance. 61.32% of samples from Kayin State carried nonsynonymous SNP mutations in \u003cem\u003ekelch13\u003c/em\u003e (Supplementary Figure 11). The major mutant alleles were P441L (15.19%), F446I (15.01%), R561H (14.02%), and G449A (7.87%). Only 2.28% of Kayin samples carry C580Y. In comparison, in the adjacent SMRU clinics, C580Y was the dominant \u003cem\u003ekelch13\u003c/em\u003e mutant allele, reaching 71.05% allele frequency in 2014 (Table S2). There were 47 IBD clusters (\u003cem\u003er\u0026nbsp;\u003c/em\u003e\u0026ge; 0.90) carrying mutant \u003cem\u003ekelch13\u003c/em\u003e alleles and 46 IBD clusters carrying wild-type \u003cem\u003ekelch13\u003c/em\u003e. We compared the size of IBD clusters carrying mutant \u003cem\u003ekelch13\u003c/em\u003e alleles with those carrying wild type \u003cem\u003ekelch13\u003c/em\u003e and found no significant difference (Figure 3C, Supplementary Figure 12). These results suggest that artemisinin selection was not the main driver for clonal expansion in Kayin State.\u003c/p\u003e\n\u003cp\u003eClonal expansion of parasite carrying \u003cem\u003ekelch13\u003c/em\u003e-R561H in 2020. Despite strong drug selection, frequencies of mutant \u003cem\u003ekelch13\u003c/em\u003e alleles remained stable between 2016 and 2019 \u003csup\u003e6\u003c/sup\u003e (Figure 3C, Supplementary Figure 12). However, in 2020 one of the \u003cem\u003ekelch13\u003c/em\u003e alleles - R561H - reached 74.2%. This clonal expansion results from near elimination of parasites from most areas in Kayin, with the exception of northern Hpapun Township where parasites carrying \u003cem\u003ekelch13\u003c/em\u003e-R561H predominate (Figure 4D). 54.8% (40/73) of samples collected between January and August 2020 before the COVID-19 pandemic lockdowns were from one single malaria post (LH-0266B) (Table S1).\u003c/p\u003e\n\u003cp\u003eThe change in \u003cem\u003ekelch13\u003c/em\u003e allele frequencies were reflected by changes in diversity in this gene and its flanking regions. We measured expected heterozygosity (\u003cem\u003eHe\u003c/em\u003e) to quantify diversity. \u003cem\u003eHe\u003c/em\u003e in \u003cem\u003ekelch13\u0026nbsp;\u003c/em\u003eremained high between 2016-2019 (\u003cem\u003eHe\u003c/em\u003e = 0.78) but dropped to 0.41 in 2020. We observed parallel reductions in flanking region diversity with a drop from 0.48 to 0.19 between 2019 and 2020 (Supplementary Figure 11D).\u003c/p\u003e\n\u003cp\u003eOrigins of \u003cem\u003ekelch13\u003c/em\u003e alleles. We reconstructed the haplotypes surrounding the \u003cem\u003ekelch13\u003c/em\u003e gene (100kb upstream and 100 kb downstream). We found a wide variety of \u003cem\u003ekelch13\u003c/em\u003e genetic backgrounds, with one or more unique haplotypes per resistance allele (Supplementary Figure 13 \u0026amp; 14). Two P441L, one F446I, one G449A and one R561H haplotypes had shared ancestry between Kayin State and SMRU clinics or other Myanmar regions. However, none of these alleles had high frequency in SMRU clinics or other Myanmar regions. Two F446I and one C580Y haplotypes were uniquely observed in Kayin State, consistent with local origin. For two \u003cem\u003eKelch13\u0026nbsp;\u003c/em\u003eresistance alleles (G449A, R561H), the wildtype \u003cem\u003ekelch13\u003c/em\u003e haplotypes on which these resistance mutations arose were sampled in the early 2000s in SMRU clinics. Of the three C580Y haplotypes identified in Kayin, two were also found in SMRU clinics, while one was unique to Kayin State. The two C580Y haplotypes shared with SMRU clinics were only found to the south of Hpapun Township and 60-120km north of the SMRU clinics (Supplementary Figure 15). None of these haplotypes shared IBD with east SE Asia C580Y haplotypes (Supplementary Figure 16).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImpact of mass drug usage on parasite population structure\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe predicted that MDA would reduce relatedness of pre and post MDA parasite populations due to clearance of the local parasite population and replacement with new parasite genotypes post-MDA, resulting in founder effects. There were three HCPC units in which sufficient parasites (n \u0026nbsp;\u0026ge; 20) were sequenced both pre and post MDA (Figure 6A). The relatedness between pre and post MDA parasites from these 3 HCPC units was significantly lower than observed between malaria parasites collected during the sampling time period from HCPC regions where MDA was not used (Figure 6B). Hence, MDA impacted parasite relatedness, consistent with efficient control of pre-MDA parasite genotypes, and post-MDA recolonization with unrelated genotypes.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe METF elimination efforts, combining community malaria posts and MDA, significantly decreased malaria case numbers in the target area (near the Myanmar-Thailand border) between 2014-2020 \u003csup\u003e8,9\u003c/sup\u003e. We described the key results from genomic surveillance during these elimination efforts, including parasite transmission patterns, population diversity and genomics, evolution of drug resistance, and genomic impacts of MDA using over 2000 whole genome sequenced \u003cem\u003eP. falciparum\u003c/em\u003e samples collected between Nov 2015 – Aug 2020.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSpatial and temporal structure of malaria populations\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eParasite sequences from the METF study region reveal extremely high levels of inbreeding and low levels of genetic variation. We found 166 genotype clusters (≥90% of the genome IBD) among the 1726 single clone samples sequenced. Hence only 10% of parasite genomes sampled are unique (R\u003csub\u003eG\u003c/sub\u003e = 0.1), and most infections show a clonal structure. In contrast, genetic richness is \u0026gt;0.94 in African populations sampled, and ranges from 0.54 to 0.89 in other SE Asian populations examined \u003csup\u003e19\u003c/sup\u003e. Furthermore, 110 of the 166 unique genomes are distributed among 9 different families (r ≥ 0.45). Parasites within these families typically carry one or two \u003cem\u003ekelch13\u003c/em\u003e alleles, that mostly likely inherited from the parents. This is clearly the case in the two families with both parents identified - family 1 (F446I and wild-type \u003cem\u003ekelch13\u003c/em\u003e) and family 7 (C580Y and wild-type \u003cem\u003ekelch13\u003c/em\u003e). Hence recombination is rare in these populations and most infections are clonally related.\u003c/p\u003e\n\u003cp\u003eWe observed strong spatial structure in the parasite population. \u0026nbsp;This is particularly clear from the distribution of unique parasite genotypes. Such parasites result from self-fertilization of male and female gametes of the same genotype and allow spatial tracing of transmission networks. That these transmission networks are clustered in space is clear visually (Figure 4A), and statistically evident from autocorrelation analyses (Figure 4B), which reveal positive correlations in relatedness between genotypes for up to 20km. The local distribution of unique genotypes indicates (i) local transmission, and very few long-distance transmission events and (ii) reintroduction of circulating genotypes from asymptomatic carriers.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eUnique parasite genotypes were long lived in this low transmission setting, with some IBD clusters observed over the complete 56-month study period. This is clear from (i) the temporal autocorrelation, which reveals positive correlation in relatedness between parasites collected 170 days apart; (ii) from the elevated relatedness observed in parasites collected from the same HCPC regions but up 3 years apart. The strong spatial and temporal sub-structure of parasite populations in Kayin State is comparable to that observed in Cambodia \u003csup\u003e20\u003c/sup\u003e and Guyana \u003csup\u003e21\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA clonal expansion of parasites carrying \u003cem\u003ekelch13\u003c/em\u003e-R561H in 2020\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs elimination approaches, genetic drift is expected to play an increasingly important role and expansions of parasite lineages may occur \u003csup\u003e22\u003c/sup\u003e. We observed a sudden increase in the frequency of IBD cluster 1 parasites carrying \u003cem\u003ekelch13\u003c/em\u003e-R561H in 2020. In this case, the increase of \u003cem\u003ekelch13\u003c/em\u003e-R561H frequency resulted from elimination of malaria from all regions of Kayin other than northern Hpapun Township (Figure 4A), where IBD cluster 1 carrying \u003cem\u003ekelch13\u003c/em\u003e-R561H is at high frequency. Despite predominating in northern Hpapun Township since 2017, the \u003cem\u003ekelch13\u003c/em\u003e-R561H didn’t spread into other areas of Kayin State. The rapid frequency increase of IBD cluster 1 in 2020 is consistent with bottlenecks and genetic drift in a \u003cem\u003eP. falciparum\u003c/em\u003e population nearing elimination. Further sampling will determine whether this parasite genotype spreads further in Kayin State and elsewhere in Myanmar and Thailand. In contrast, Wasakul et al \u003csup\u003e22\u003c/sup\u003e describe a classic outbreak driven by a selective sweep in Laos, where a lineage carrying \u003cem\u003ekelch13\u003c/em\u003e-R539H (named LAA1) rose from a low frequency to replace the previously dominant KEL1/PLA1 (C580Y) population.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMDA impacts parasite population structure\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTwo control measures were used in Kayin State by METF: malaria posts (early diagnosis and community case management), and regional MDA in malaria hotspots \u003csup\u003e8\u003c/sup\u003e. The combination of these approaches significantly decreased malaria incidence \u003csup\u003e8,9\u003c/sup\u003e. Encouragingly, there was minimal evidence of selection for drug-resistance parasites through these elimination efforts which agrees with observations by Imwong et al \u003csup\u003e23\u003c/sup\u003e, McLean et al \u003csup\u003e24\u003c/sup\u003e and Thu et al \u003csup\u003e6\u003c/sup\u003e. At the genomic level, these combined control measures reduced parasite effective population size (Figure 5).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis study provided an opportunity to examine the impact of one component of this elimination strategy - MDA - on parasite population structure. MDA is expected to generate bottlenecks between pre and post MDA malaria populations, because reservoirs of asymptomatic malaria are removed. We therefore expect to see founder effects resulting from newly colonizing parasites and large divergence between pre and post MDA populations, when compared to populations with no MDA. Our power to detect an impact was limited because most parasites sequenced were collected after MDA: only three HCPC units had sufficient numbers of infection sampled both pre and post MDA to examine this hypothesis. Nevertheless, as predicted, we saw lower relatedness between pre and post treatment parasites in these three HCPC regions than in control regions where MDA was not implemented. These results provide genomic evidence for the effectiveness of MDA in removing local parasite populations through effective clearance of both asymptomatic and symptomatic parasite infections.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGenomic metrics for assessing transmission intensity\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGenetic metrics, such as proportion of multiple clone infections can provide useful metrics for assessing control efforts \u003csup\u003e18,25,26\u003c/sup\u003e. Such metrics are particularly useful in low transmission regions, where high prevalence of low-density asymptomatic infections complicates assessment of transmission using standard epidemiological methods. However, genetic metrics perform poorly when transmission levels are extremely low. In Senegal \u003csup\u003e25,26\u003c/sup\u003e, complexity of infection provided an unreliable metric for evaluating transmission when transmission is incidence \u0026lt; 100 cases per 1000 person-years. Consistent with this, we observed that both R\u003csub\u003eG\u003c/sub\u003e and proportion of Multiple infections worked poorly in Kayin where incidence ranged from 1 - 39 cases per 1000 person years. However, we found that another metric, effective population size (N\u003csub\u003ee\u003c/sub\u003e) calculated from LD between unlinked markers, showed significant decline in both the Kayin State parasite population and from SMRU clinics. We suggest that N\u003csub\u003ee\u003c/sub\u003e may be a useful genomic indicator of transmission dynamics, particularly in parasite populations in which transmission has been reduced to near elimination levels. N\u003csub\u003ee\u003c/sub\u003e is typically used in conservation biology to assess viability in endangered populations of animals and plants. Our results suggest that this metric may also have utility for assessing whether parasite populations are approaching local extinction.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNo evidence for long-distance gene flow between Kayin State and Eastern SE Asian countries\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe current frontline treatments for \u003cem\u003eP. falciparum\u003c/em\u003e malaria parasites have been failing in east SE Asia \u003csup\u003e23,27,28\u003c/sup\u003e, due to the spread of the multidrug-resistance parasites carrying \u003cem\u003ekelch13\u003c/em\u003e-C580Y mutation and \u003cem\u003eplasmepsin\u003c/em\u003e 2 amplifications, named KEL1/PLA1. The KEL1/PLA1 lineage was first detected in Cambodia as the DHA-piperaquine was heavily used. When Cambodia withdraw DHA-piperaquine and adopted to artesunate–mefloquine, KEL1/PLA1 subgroups with acquired \u003cem\u003epfcrt\u003c/em\u003e mutations conferring piperaquine resistance rapidly spread to other ESEA countries, such as Laos and Vietnam \u003csup\u003e27\u003c/sup\u003e. There is a concern that this parasite lineage will further spread to west SE Asia, which has the majority of malaria cases in SE Asia \u003csup\u003e24\u003c/sup\u003e and where DHA-piperaquine is the frontline treatment for \u003cem\u003eP. falciparum\u003c/em\u003e. Two lines of evidence suggest minimal geneflow between east and west SE Asia. First, we did not detect \u003cem\u003epfcrt\u003c/em\u003e mutations associated with piperaquine resistance on Thailand-Myanmar border or in the Kayin State sampling sites. Second, examination of genome-wide IBD sharing among 3718 infections (1458 unique genomes) revealed no recent recombination or clonal transmission between east and west SE Asia (Supplementary Figure 9).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOrigins of kelch13 resistance alleles\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ekelch13\u003c/em\u003e mutations conferring artemisinin resistance are established in both east and west SE Asia. C580Y is the major mutation in regions other than northern Myanmar, where F446I predominates \u003csup\u003e23\u003c/sup\u003e. In contrast, the dominant \u003cem\u003ekelch13\u003c/em\u003e mutations in the Kayin State include P441L, F446I, R561H, and G449A, depending on the location (Figure 4). While the majority of infections from nearby SMRU clinics carry C580Y (71.05% in 2014), the C580Y frequency in Kayin State was only 2.28%. The most widespread F446I haplotypes in Kayin State originated independently from the dominant F446I haplotype in northern Myanmar.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhat factors lead to the patterns of artemisinin resistance evolution seenin Kayin? Longitudinal studies in both Cambodia and from SMRU clinics have revealed that multiple independent \u003cem\u003ekelch13\u003c/em\u003e mutations emerged and spread initially (soft selective sweeps). Single \u003cem\u003ekelch13\u003c/em\u003e genotypes (typically \u003cem\u003ekelch13\u003c/em\u003e-C580Y) eventually outcompete other lineages leading to hard selective sweeps \u003csup\u003e2,20,23,27\u003c/sup\u003e. In contrast, we found limited evidence that strong drug selection drove drug resistance evolution \u0026nbsp;in Kayin State: (i) we found no significant increase in \u003cem\u003ekelch13\u003c/em\u003e mutant allele frequencies before 2020 (Figure 3C, Supplementary Figure 11) \u003csup\u003e6\u003c/sup\u003e; (ii) the size of clonal clusters was not significantly different when comparing \u003cem\u003ekelch13\u0026nbsp;\u003c/em\u003ewildtype and mutant parasites (Figure 3C, Supplementary Figure 12). The small effective population size of malaria parasite populations may contribute to the patterns observed, because selection is inefficient when population sizes are small and genetic drift is enhanced \u003csup\u003e29\u003c/sup\u003e. The initial effective population size of malaria parasites in the Kayin State dataset was much smaller (Ne = 11.5 to 26.6) compared to SMRU clinics (15.5 – infinite) (Figure 2, Figure 5). Other factors that may also limit the impact of drug selection. Human population movements were more limited in Kayin compared to nearby SMRU clinics, especially in Northern Hpapun Township where human movement is limited by difficult terrain, the heavily militarized landscape, and a lack of year-round roads \u003csup\u003e7,30\u003c/sup\u003e, which can hinder transmission of resistance alleles. Similarly, low levels of recombination in Kayin State limits the rate of formation of new multi-locus parasite genotypes. The small parasite population size, limited population movement, and minimal recombination enhance the role of genetic drift rather than selection in determining drug resistance evolution in the Kayin State region. Our results, and those from other studies \u003csup\u003e23\u003c/sup\u003e illustrate how genetic drift can result in rapid stochastic changes in parasite population genomics and drug resistance status in small parasite populations close to elimination.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis study had several limitations: (i) We analyzed malaria genomes collected from 2015 onwards. However, control efforts began earlier than this in 2014. Hence, we were unable to examine malaria population structure and diversity prior to initiation of control efforts. (ii) Use of finger prick blood samples and whole genome amplification resulted in bias towards sequencing high parasitemia infections. (iii) we were unable to score copy number variants, in genes such as \u003cem\u003ePlasmepsin\u003c/em\u003e II/III, associated with piperaquine resistance from whole genome amplified DNA. However, the sustained sampling of a high proportion of blood spots collected over a 5-year period provides a unique dataset for examining impact of malaria control efforts on parasite population structure and resistance evolution.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eStudy area and sample origins\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe samples for this analysis were collected during routine diagnosis and treatment efforts in Kayin State as part of the METF malaria elimination effort led by the Shoklo Malaria Research Unit (SMRU, based on the Thailand-Myanmar border) (Figure 1, Supplementary Figure 1\u0026amp; 2). This METF project was established in 2014 and utilized two primary \u003cem\u003eP.\u003c/em\u003e \u003cem\u003efalciparum\u003c/em\u003e-focused interventions: the establishment of a large network of community-based malaria diagnosis and treatment posts (MPs), and targeted MDA in communities determined to have a high prevalence of asymptomatic \u003cem\u003eP.\u003c/em\u003e \u003cem\u003efalciparum\u0026nbsp;\u003c/em\u003einfections. The MPs were stocked with filter papers (Whatman 3mm blotting paper) and were asked to collect dried blood spots (DBSs) from finger prick blood samples for patients with rapid diagnostic test (RDT) confirmed \u003cem\u003eP. falciparum\u003c/em\u003e infection. Each DBS sample is linked to the\u0026nbsp;MP from which it originated, and all MPs have geographic references (latitude and longitude). 5014 DBS samples were collected between November 2015 and August 2020 (Table S1). The DBS samples were then transported to SMRU and subsequently shipped to the Texas Biomedical Research Institute (in the U.S.A.) for molecular analyses.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSequencing library preparation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe extracted DNA from the dried blood spots and enriched parasite genomes using selective whole genome amplification (sWGA) to following Li et al\u0026nbsp;\u003csup\u003e31\u003c/sup\u003e and Oyola et al \u003csup\u003e32\u003c/sup\u003e. We extracted and purified genomic DNA using QIAamp\u0026reg; 96 DNA Blood Kit or QIAamp DNA Mini Kit, following the instruction manual for DNA isolation from dried blood spots. The DNA was then eluted in 100ul of 10mM Tris-HCl (pH 8.0-8.5) buffer. We used real-time quantitative PCR (qPCR) to estimate the numbers of parasite genomes in each DNA sample as described in Li et al\u0026nbsp;\u003csup\u003e31\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor samples with more than 200 copies of parasite genome per ul, we used selective whole genome amplification (sWGA) to enrich parasite DNA. sWGA reactions were performed following Oyola et al \u003csup\u003e32\u003c/sup\u003e. Each 25 \u0026mu;l reaction contained at least 1000 copies of \u003cem\u003ePlasmodium\u003c/em\u003e DNA, 1\u0026times; BSA (New England Biolabs), 1 mM dNTPs (New England Biolabs), 3.5 \u0026mu;M of each amplification primer, 1\u0026times; Phi29 reaction buffer (New England Biolabs), and 15 units of Phi29 polymerase (New England Biolabs). We used a PCR machine (SimpliAmp, Applied Biosystems) programmed to run a \u0026ldquo;stepdown\u0026rdquo; protocol: 35 \u0026deg;C for 10 min, 34 \u0026deg;C for 10 min, 33 \u0026deg;C for 10 min, 32 \u0026deg;C for 10 min, 31 \u0026deg;C for 10 min, 30 \u0026deg;C for 6 h then heating at 65 \u0026deg;C for 10 min to inactivate the enzymes prior to cooling to 4 \u0026deg;C. Sample were cleaned with AMPure XP Beads (Beckman Coulter), at a 1:1 ratio.\u003c/p\u003e\n\u003cp\u003eWe used the Quant-iT\u0026trade; PicoGreen\u0026reg; Assay (Invitrogen) to determine the total amount of sWGA product, and quantified the proportion of\u003cem\u003e\u0026nbsp;Plasmodium\u0026nbsp;\u003c/em\u003eDNA by qPCR. Only sWGA products with more than 50% DNA from \u003cem\u003ePlasmodium\u003c/em\u003e were used for further library preparation and Illumina sequencing. We constructed PCR-free next generation sequencing libraries using 300ng sWGA product following the KAPA HyperPlus Kit protocol. All libraries were sequenced to an average coverage of 60\u0026times; using Illumina Hiseq X or Novaseq sequencers.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWhole-genome sequencing data generation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe individually mapped whole-genome sequencing reads for each library against the \u003cem\u003eP. falciparum\u003c/em\u003e 3D7 reference genome (PlasmoDB, release 46) and human GRCh38 reference genome, using the alignment algorithm BWA mem (http://bio-bwa.sourceforge.net/) under the default parameters. The resulting alignments were then converted to SAM format, sorted to BAM format, and deduplicated using \u003cem\u003epicard\u003c/em\u003e tools v2.0.1 (http://broadinstitute.github.io/picard/). Reads mapping to the human genome were discarded before genotyping.\u003c/p\u003e\n\u003cp\u003eWe used Genome Analysis Toolkit GATK v3.7 (https://software.broadinstitute.org/gatk/) to recalibrate the base quality score based on a set of verified known variants \u003csup\u003e33\u003c/sup\u003e. We called variants for each sample using HaplotypeCaller, and calls from every 100 samples were merged using \u003cem\u003eCombineGVCFs\u003c/em\u003e with default parameters. Variants were further called at all sample-level using \u003cem\u003eGenotypeGVCFs\u003c/em\u003e, with parameters: --max_alternate_alleles 6 --variant_index_type LINEAR --variant_index_parameter 128000 --sample_ploidy 2 -nt 20.\u003c/p\u003e\n\u003cp\u003eThe recalibrated variant quality scores (VQSR) were calculated by comparing the raw variant distribution with the known and verified \u003cem\u003ePlasmodium\u003c/em\u003e variant dataset. SNPs and indes with VQSR less than 1 or located outside of the core genome regions (21 Mb, defined in \u003csup\u003e33\u003c/sup\u003e) were removed from further analysis. Samples with less than 50% of the core genome callable were also excluded from further analysis. Only biallelic SNPs that pass all the quality filter were used, unless otherwise specified. The final variants in VCF format were annotated at functional effect to genes and proteins using snpEff v4.3 (https://pcingola.github.io/SnpEff/) with 3D7 (PlasmoDB, release 46) as reference.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe initially identified 1,302,006 single-nucleotide polymorphisms (SNPs) and 703,138 indels (Figure 1C). We removed 343 samples with \u0026gt; 20% genotypes missing. We then filtered the SNP calls following a \u0026ldquo;stringent\u0026rdquo; filtering method \u003csup\u003e34\u003c/sup\u003e, to generate a final list of 447,435 high-quality, biallelic, core-genome located (defined in \u003csup\u003e33\u003c/sup\u003e) SNPs. To analyze complexity of infection and population structure, we further removed SNPs that were genotyped in less than 50% of samples or with minor allele frequency (MAF) \u0026lt; 0.05.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eComplexity of infection\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe measured multiplicity of \u003cem\u003eP. falciparum\u003c/em\u003e infections using the within-infection \u003cem\u003eF\u003csub\u003ews\u0026nbsp;\u003c/sub\u003e\u003c/em\u003efixation index \u003csup\u003e35\u003c/sup\u003e. Samples with \u003cem\u003eF\u003csub\u003eWS\u003c/sub\u003e\u003c/em\u003e \u0026gt; 0.9 were assumed to come from single-genotype infections for samples from Kayin State. Allele frequencies across the genome were plotted and manually inspected to detect further possible complex infections.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRelationships among parasite genotypes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe used relatedness - \u003cem\u003er\u003c/em\u003e, defined as the fraction of the genome that is identical-by-descent (IBD) between a pair of individuals \u003csup\u003e36,37\u003c/sup\u003e - to estimate parasite relationships. Based on the distribution of relatedness among F1 progeny from malaria parasite genetic crosses (Supplementary Figure 3), we assume that parasites are genetically related if \u0026ge; 25% of their genome is identical (\u003cem\u003er\u0026nbsp;\u003c/em\u003e\u0026ge; 0.25); parasites are closely related (such as siblings or parent and progeny) if their relatedness is greater than 45% (\u003cem\u003er\u003c/em\u003e \u0026ge; 0.45). We considered samples to be clonal if their relatedness is over 90% (\u003cem\u003er\u003c/em\u003e \u0026ge; 0.90). We visualized relatedness among samples using the R package \u003cem\u003epheatmap\u003c/em\u003e and the \u003cem\u003eCytoscape\u003c/em\u003e software. We also examined the recombination patterns between closely related parasites and plotted shared IBD regions between estimated parents and progeny using \u003cem\u003ekaryoploteR\u003c/em\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSurveillance of \u003cem\u003ekelch13\u003c/em\u003e haplotypes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe extracted SNPs distributed within 100 kb upstream and 100 kb downstream of the \u003cem\u003ekelch13\u003c/em\u003e gene. We measured expected heterozygosity (\u003cem\u003eHe\u003c/em\u003e) at the \u003cem\u003ekelch13\u003c/em\u003e locus by treating \u003cem\u003ekelch13\u003c/em\u003e as a single locus with multiple alleles. We also measured \u003cem\u003eHe\u003c/em\u003e over the 200kb \u003cem\u003ekelch13\u003c/em\u003e haplotype region. To compare the relationships between different \u003cem\u003ekelch13\u003c/em\u003e haplotypes, we measured pairwise IBD sharing among all \u003cem\u003ekelch13\u003c/em\u003e haplotypes. We assume that haplotypes with IBD sharing \u0026ge; 0.90 originated from the same mutational event; that when 0.35 \u0026le; IBD \u0026lt; 0.90, there was a one least recombination event to break the original haplotype; and when IBD \u0026lt; 0.35, these haplotypes have emerged independently.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eComparisons of malaria parasite populations\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe compared the Kayin State parasite population with other world-wide malaria parasite populations (Figure 1, Table S2). The SMRU clinics are located around Mae Sot, in Tak Province along the international Thailand-Myanmar border. We used \u0026ldquo;other Myanmar\u0026rdquo; to represent sampling sites in Myanmar but not from Kayin State. West SE Asia population includes samples from Kayin State, SMRU clinics and other Myanmar regions, while east SE Asia population includes Cambodia, Viet Nam, and Laos.\u003c/p\u003e\n\u003cp\u003eWe merged raw SNP genotypes from the Kayin dataset with those from MalariaGEN \u003cem\u003eP. falciparum\u003c/em\u003e Community Project \u003csup\u003e19\u003c/sup\u003e (release 6). We performed \u0026ldquo;stringent\u0026rdquo; filtration as described above, and selected loci with minor allele frequency \u0026gt; 0.05. We calculated genetic richness [R\u003csub\u003eG\u003c/sub\u003e = (G-1)/(S-1)] \u003csup\u003e38,39\u003c/sup\u003e to quantify the richness of clonal parasites in each population, where G is the number of unique genomes, and S is the total number of single genotype infected samples. For samples with relatedness \u0026gt; 0.9, only one representative sample per population with the highest genotype rate was selected and used for further analysis (Table S2). We pruned SNPs for linkage disequilibrium (LD) and generated a pairwise genetic distance matrix using PLINK with default parameters. We conducted principal component analyses (PCA) and ADMIXTURE analyses based on the pruned genotypes and distance matrix. We measured the proportion of pairs IBD across the genome within populations following the scripts in \u003cem\u003eisoRelate\u003c/em\u003e \u003csup\u003e40\u003c/sup\u003e. We estimated effective population size (\u003cem\u003eNe\u003c/em\u003e) based on patterns of LD at unlinked loci, using methods implemented in \u003cem\u003eNeEstimator\u0026nbsp;\u003c/em\u003ev2.0 \u003csup\u003e41\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll statistical analysis was performed using R version 4.1.3. For pairwise comparisons between groups, we used Welch Two Sample T-test. We measured correlations between parasite genetic relatedness and geographic distance or time using the Mantel statistic using the \u003cem\u003emantel\u0026nbsp;\u003c/em\u003efunction in the \u003cem\u003evegan\u003c/em\u003e package. \u003cem\u003ep\u003c/em\u003e\u0026lt;0.05 was considered statistically significant. We used hierarchical clustering on principal components (HCPC) following scripts in \u003cem\u003eFactoMineR\u003c/em\u003e \u003csup\u003e42\u003c/sup\u003e to divide the 283 malaria posts with samples sequenced into 50 HCPC regions based on latitude and longitude. We then compared parasite relatedness within and between HCPC regions for parasites collected in the same year, between 1-2, 2-3 and 3-4 years apart. We compared relatedness between parasites collected from HCPC regions 6 months before and 6 months after MDA. As controls, we examined relatedness of parasites collected from HCPC regions where MDA was not used during the same time windows.\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by National Institutes of Health (NIH) grant R37 AI048071 (to TJCA) and P01 AI127338. Work at Texas Biomedical Research Institute was conducted in facilities constructed with support from Research Facilities Improvement Program grant C06 RR013556 from the National Center for Research Resources. SMRU is part of the Mahidol Oxford University Research Unit supported by the Wellcome Trust of Great Britain. The malaria elimination program (Malaria Elimination Task Force, METF) in Kayin State, Myanmar is supported by the Regional Artemisinin Initiative (Global Fund to Fight AIDS, Tuberculosis and Malaria) and the Bill and Melinda Gates Foundation (OPP1117507). The authors would like to acknowledge the contribution from all member of METF and SMRU, collaborators, and colleagues who have supported the elimination program.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eXL, FN, and TJCA conceived and designed the study. JL, AMT, GD, DMP, KML, KS and FN coordinated sample and data collection. XL, GAA, and AR processed samples, and generated genomic data. XL analyzed and interpreted the sequencing data, with input from TJCA, DMP, JL and FN. KML, KS, JL, DMP, and FN were involved in the management and coordination of the genetic surveillance project. XL and TJCA wrote the initial manuscript. XL, DMP, JL, FN and TJCA revised the manuscript. XL, DMP, FN and TJCA accessed and verified all the data. All authors provided critical revision of the manuscript. All authors had full access to all the data in the study and had final responsibility for the decision to submit for publication. TJCA and FN contributed equally.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAuthors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAdditional Information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSupplementary Information is available for this paper.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRaw sequencing data for the 2270 sequenced samples collected by the Malaria Elimination Task Force project from Myanmar used in the present analysis have been submitted to the NABI Sequence Read Archive (SRA, https://www.ncbi.nlm.nih.gov/sra) under the project number of PRJNA864839. The analysis code and data matrices (genetic distances, geographic distances and temporal distances) are available at: https://github.com/emilyli0325/Malaria-genomics-in-Eastern-Myanmar. This publication also uses data from the MalariaGEN \u003cem\u003eP falciparum\u0026nbsp;\u003c/em\u003eCommunity Project \u003csup\u003e19\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNational Institutes of Health grant R37 AI048071 (TJCA);\u003c/p\u003e\n\u003cp\u003eNational Institutes of Health grant 2P01AI127338 (TJCA: Core B PI);\u003c/p\u003e\n\u003cp\u003eThe malaria elimination program (Malaria Elimination Task Force, METF) in Kayin State, Myanmar is supported by the Regional Artemisinin Initiative (Global Fund to Fight AIDS, Tuberculosis and Malaria) and the Bill and Melinda Gates Foundation (OPP1117507) (FN);\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor the purpose of open access, the author has applied a CC BY public copyright license to any Author Accepted Manuscript version arising from this submission.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eVenkatesan, P. The 2023 WHO World malaria report. \u003cem\u003eLancet Microbe\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, e214, doi:10.1016/S2666-5247(24)00016-8 (2024).\u003c/li\u003e\n\u003cli\u003eAnderson, T. J.\u003cem\u003e et al.\u003c/em\u003e Population Parameters Underlying an Ongoing Soft Sweep in Southeast Asian Malaria Parasites. \u003cem\u003eMol Biol Evol\u003c/em\u003e \u003cstrong\u003e34\u003c/strong\u003e, 131-144, doi:10.1093/molbev/msw228 (2017).\u003c/li\u003e\n\u003cli\u003eWellems, T. E. \u0026amp; Plowe, C. V. Chloroquine-resistant malaria. \u003cem\u003eJ Infect Dis\u003c/em\u003e \u003cstrong\u003e184\u003c/strong\u003e, 770-776, doi:10.1086/322858 (2001).\u003c/li\u003e\n\u003cli\u003eImwong, M.\u003cem\u003e et al.\u003c/em\u003e Numerical Distributions of Parasite Densities During Asymptomatic Malaria. \u003cem\u003eJ Infect Dis\u003c/em\u003e \u003cstrong\u003e213\u003c/strong\u003e, 1322-1329, doi:10.1093/infdis/jiv596 (2016).\u003c/li\u003e\n\u003cli\u003eWhite, N. J. Does antimalarial mass drug administration increase or decrease the risk of resistance? \u003cem\u003eLancet Infect Dis\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, e15-e20, doi:10.1016/S1473-3099(16)30269-9 (2017).\u003c/li\u003e\n\u003cli\u003eThu, A. M.\u003cem\u003e et al.\u003c/em\u003e Molecular markers of artemisinin resistance during falciparum malaria elimination in Eastern Myanmar. \u003cem\u003eMalar J\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 138, doi:10.1186/s12936-024-04955-6 (2024).\u003c/li\u003e\n\u003cli\u003eParker, D. M.\u003cem\u003e et al.\u003c/em\u003e Scale up of a Plasmodium falciparum elimination program and surveillance system in Kayin State, Myanmar. \u003cem\u003eWellcome Open Res\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 98, doi:10.12688/wellcomeopenres.12741.2 (2017).\u003c/li\u003e\n\u003cli\u003eLandier, J.\u003cem\u003e et al.\u003c/em\u003e Effect of generalised access to early diagnosis and treatment and targeted mass drug administration on Plasmodium falciparum malaria in Eastern Myanmar: an observational study of a regional elimination programme. \u003cem\u003eLancet\u003c/em\u003e \u003cstrong\u003e391\u003c/strong\u003e, 1916-1926, doi:10.1016/S0140-6736(18)30792-X (2018).\u003c/li\u003e\n\u003cli\u003eLegendre, E.\u003cem\u003e et al.\u003c/em\u003e \u0026apos;Forest malaria\u0026apos; in Myanmar? Tracking transmission landscapes in a diversity of environments. \u003cem\u003eParasit Vectors\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 324, doi:10.1186/s13071-023-05915-w (2023).\u003c/li\u003e\n\u003cli\u003eShomuyiwa, D. O.\u003cem\u003e et al.\u003c/em\u003e Cabo Verde\u0026apos;s malaria-free certification: A blueprint for eradicating malaria in Africa. \u003cem\u003eJ Taibah Univ Med Sci\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 534-536, doi:10.1016/j.jtumed.2024.04.001 (2024).\u003c/li\u003e\n\u003cli\u003eYoung, N. W.\u003cem\u003e et al.\u003c/em\u003e High frequency of artemisinin partial resistance mutations in the great lake region revealed through rapid pooled deep sequencing. \u003cem\u003emedRxiv\u003c/em\u003e, doi:10.1101/2024.04.29.24306442 (2024).\u003c/li\u003e\n\u003cli\u003eRosenthal, P. J.\u003cem\u003e et al.\u003c/em\u003e The emergence of artemisinin partial resistance in Africa: how do we respond? \u003cem\u003eLancet Infect Dis\u003c/em\u003e, doi:10.1016/S1473-3099(24)00141-5 (2024).\u003c/li\u003e\n\u003cli\u003eMalariaGen\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, doi:10.12688/wellcomeopenres.18681.1 (2023).\u003c/li\u003e\n\u003cli\u003eBoonyalai, N.\u003cem\u003e et al.\u003c/em\u003e Piperaquine resistant Cambodian Plasmodium falciparum clinical isolates: in vitro genotypic and phenotypic characterization. \u003cem\u003eMalar J\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 269, doi:10.1186/s12936-020-03339-w (2020).\u003c/li\u003e\n\u003cli\u003eSmall-Saunders, J. L.\u003cem\u003e et al.\u003c/em\u003e Evidence for the early emergence of piperaquine-resistant Plasmodium falciparum malaria and modeling strategies to mitigate resistance. \u003cem\u003ePLoS Pathog\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, e1010278, doi:10.1371/journal.ppat.1010278 (2022).\u003c/li\u003e\n\u003cli\u003eRoss, L. S.\u003cem\u003e et al.\u003c/em\u003e Emerging Southeast Asian PfCRT mutations confer Plasmodium falciparum resistance to the first-line antimalarial piperaquine. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 3314, doi:10.1038/s41467-018-05652-0 (2018).\u003c/li\u003e\n\u003cli\u003eKane, J.\u003cem\u003e et al.\u003c/em\u003e A Plasmodium falciparum genetic cross reveals the contributions of pfcrt and plasmepsin II/III to piperaquine drug resistance. \u003cem\u003ebioRxiv\u003c/em\u003e, doi:10.1101/2023.06.06.543862 (2023).\u003c/li\u003e\n\u003cli\u003eNkhoma, S. C.\u003cem\u003e et al.\u003c/em\u003e Population genetic correlates of declining transmission in a human pathogen. \u003cem\u003eMol Ecol\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 273-285, doi:10.1111/mec.12099 (2013).\u003c/li\u003e\n\u003cli\u003eMalariaGen\u003cem\u003e et al.\u003c/em\u003e An open dataset of Plasmodium falciparum genome variation in 7,000 worldwide samples. \u003cem\u003eWellcome Open Res\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 42, doi:10.12688/wellcomeopenres.16168.2 (2021).\u003c/li\u003e\n\u003cli\u003eMiotto, O.\u003cem\u003e et al.\u003c/em\u003e Multiple populations of artemisinin-resistant Plasmodium falciparum in Cambodia. \u003cem\u003eNature genetics\u003c/em\u003e \u003cstrong\u003e45\u003c/strong\u003e, 648-655, doi:10.1038/ng.2624 (2013).\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, doi:10.1371/journal.ppat.1012013 (2024).\u003c/li\u003e\n\u003cli\u003eWasakul, V.\u003cem\u003e et al.\u003c/em\u003e Malaria outbreak in Laos driven by a selective sweep for Plasmodium falciparum kelch13 R539T mutants: a genetic epidemiology analysis. \u003cem\u003eLancet Infect Dis\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 568-577, doi:10.1016/S1473-3099(22)00697-1 (2023).\u003c/li\u003e\n\u003cli\u003eImwong, M.\u003cem\u003e et al.\u003c/em\u003e Molecular epidemiology of resistance to antimalarial drugs in the Greater Mekong subregion: an observational study. \u003cem\u003eLancet Infect Dis\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 1470-1480, doi:10.1016/S1473-3099(20)30228-0 (2020).\u003c/li\u003e\n\u003cli\u003eMcLean, A. R. D.\u003cem\u003e et al.\u003c/em\u003e Mass drug administration for the acceleration of malaria elimination in a region of Myanmar with artemisinin-resistant falciparum malaria: a cluster-randomised trial. \u003cem\u003eLancet Infect Dis\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 1579-1589, doi:10.1016/S1473-3099(20)30997-X (2021).\u003c/li\u003e\n\u003cli\u003eWong, W.\u003cem\u003e et al.\u003c/em\u003e Evaluating the performance of Plasmodium falciparum genetic metrics for inferring National Malaria Control Programme reported incidence in Senegal. \u003cem\u003eMalar J\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 68, doi:10.1186/s12936-024-04897-z (2024).\u003c/li\u003e\n\u003cli\u003eSchaffner, S. F.\u003cem\u003e et al.\u003c/em\u003e Malaria surveillance reveals parasite relatedness, signatures of selection, and correlates of transmission across Senegal. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 7268, doi:10.1038/s41467-023-43087-4 (2023).\u003c/li\u003e\n\u003cli\u003eHamilton, W. L.\u003cem\u003e et al.\u003c/em\u003e Evolution and expansion of multidrug-resistant malaria in southeast Asia: a genomic epidemiology study. \u003cem\u003eLancet Infect Dis\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 943-951, doi:10.1016/S1473-3099(19)30392-5 (2019).\u003c/li\u003e\n\u003cli\u003eAmaratunga, C.\u003cem\u003e et al.\u003c/em\u003e Dihydroartemisinin-piperaquine resistance in Plasmodium falciparum malaria in Cambodia: a multisite prospective cohort study. \u003cem\u003eLancet Infect Dis\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 357-365, doi:10.1016/S1473-3099(15)00487-9 (2016).\u003c/li\u003e\n\u003cli\u003eBarton, N. H. Natural selection and random genetic drift as causes of evolution on islands. \u003cem\u003ePhilos Trans R Soc Lond B Biol Sci\u003c/em\u003e \u003cstrong\u003e351\u003c/strong\u003e, 785-794; discussion 795, doi:10.1098/rstb.1996.0073 (1996).\u003c/li\u003e\n\u003cli\u003eParker, D. M., Carrara, V. I., Pukrittayakamee, S., McGready, R. \u0026amp; Nosten, F. H. Malaria ecology along the Thailand-Myanmar border. \u003cem\u003eMalar J\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 388, doi:10.1186/s12936-015-0921-y (2015).\u003c/li\u003e\n\u003cli\u003eLi, X.\u003cem\u003e et al.\u003c/em\u003e Genetic mapping of fitness determinants across the malaria parasite Plasmodium falciparum life cycle. \u003cem\u003ePLoS Genetics\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, e-1008453 (2019).\u003c/li\u003e\n\u003cli\u003eOyola, S. O.\u003cem\u003e et al.\u003c/em\u003e Whole genome sequencing of Plasmodium falciparum from dried blood spots using selective whole genome amplification. \u003cem\u003eMalaria journal\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 597 (2016).\u003c/li\u003e\n\u003cli\u003eMiles, A.\u003cem\u003e et al.\u003c/em\u003e Indels, structural variation, and recombination drive genomic diversity in Plasmodium falciparum. \u003cem\u003eGenome research\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 1288-1299 (2016).\u003c/li\u003e\n\u003cli\u003eMcDew-White, M.\u003cem\u003e et al.\u003c/em\u003e Mode and Tempo of Microsatellite Length Change in a Malaria Parasite Mutation Accumulation Experiment. \u003cem\u003eGenome Biol Evol\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 1971-1985, doi:10.1093/gbe/evz140 (2019).\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).\u003c/li\u003e\n\u003cli\u003eBrown, T. S., Arogbokun, O., Buckee, C. O. \u0026amp; Chang, H. H. Distinguishing gene flow between malaria parasite populations. \u003cem\u003ePLoS Genet\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, e1009335, doi:10.1371/journal.pgen.1009335 (2021).\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, 1-4 (2018).\u003c/li\u003e\n\u003cli\u003eEckert, C. G., Dorken, M. E. \u0026amp; Mitchell, S. A. Loss of Sex in Clonal Populations of a Flowering Plant, Decodon Verticillatus (Lythraceae). \u003cem\u003eEvolution\u003c/em\u003e \u003cstrong\u003e53\u003c/strong\u003e, 1079-1092, doi:10.1111/j.1558-5646.1999.tb04523.x (1999).\u003c/li\u003e\n\u003cli\u003eEcheverry, D. F.\u003cem\u003e et al.\u003c/em\u003e Long term persistence of clonal malaria parasite Plasmodium falciparum lineages in the Colombian Pacific region. \u003cem\u003eBMC Genet\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 2, doi:10.1186/1471-2156-14-2 (2013).\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 Genet\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, e1007279, doi:10.1371/journal.pgen.1007279 (2018).\u003c/li\u003e\n\u003cli\u003eDo, C.\u003cem\u003e et al.\u003c/em\u003e NeEstimator v2: re-implementation of software for the estimation of contemporary effective population size (Ne ) from genetic data. \u003cem\u003eMolecular ecology resources\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 209-214, doi:10.1111/1755-0998.12157 (2014).\u003c/li\u003e\n\u003cli\u003eL\u0026ecirc;, S., Josse, J. \u0026amp; Husson, F. FactoMineR: an R package for multivariate analysis. \u003cem\u003eJournal of statistical software\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 1-18, doi:10.18637/jss.v025.i01 (2008).\u003c/li\u003e\n\u003c/ol\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-6875020/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6875020/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The malaria elimination program in Kayin State (Myanmar) utilizes malaria posts for rapid detection and treatment together with mass drug administration (MDA) in high transmission villages, and has reduced transmission by 97%. We examined the impact of control on parasite genomic parameters, using 2270 genome sequenced Plasmodium falciparum infections from 283 malaria posts, sampled over 58-months (2015 - 2020). Parasites were genetically depauperate: 1726 single-genotype infections comprised 166 unique genomes (≥90% IBD), while nine families (≥45% IBD) accounted for 62.5% of parasites sampled. Parasite effective population size decreased over the study period, but there was minimal change in artemisinin resistance alleles until 2020, when just one predominant genotype (carrying kelch13-R561H) remained. We observed sustained localized transmission of unique parasite genotypes revealing transmission chains: this resulted in positive correlations in parasite relatedness for ≤20 km. MDA resulted in parasite founder effects, providing genomic evidence for the efficacy of this control tool. These results reveal changes in population structure driven by control, and rapid shifts in allele frequency in a parasite population close to elimination.","manuscriptTitle":"Impact of intensive control on malaria population genomics in Eastern Myanmar","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-26 03:32:08","doi":"10.21203/rs.3.rs-6875020/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-microbiology","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"nmicrobiol","sideBox":"Learn more about [Nature Microbiology](http://www.nature.com/nmicrobiol/)","snPcode":"","submissionUrl":"","title":"Nature Microbiology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"370d1290-1cfc-4218-bba9-29544d802813","owner":[],"postedDate":"June 26th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":50585218,"name":"Biological sciences/Microbiology/Parasitology/Parasite genomics"},{"id":50585219,"name":"Health sciences/Diseases/Infectious diseases/Malaria"}],"tags":[],"updatedAt":"2026-04-14T07:12:42+00:00","versionOfRecord":{"articleIdentity":"rs-6875020","link":"https://doi.org/10.1038/s41564-026-02327-1","journal":{"identity":"nature-microbiology","isVorOnly":false,"title":"Nature Microbiology"},"publishedOn":"2026-04-13 04:00:00","publishedOnDateReadable":"April 13th, 2026"},"versionCreatedAt":"2025-06-26 03:32:08","video":"","vorDoi":"10.1038/s41564-026-02327-1","vorDoiUrl":"https://doi.org/10.1038/s41564-026-02327-1","workflowStages":[]},"version":"v1","identity":"rs-6875020","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6875020","identity":"rs-6875020","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}