{"paper_id":"1f22dbed-71c3-444c-a332-eb6174d4d61e","body_text":"1 \n \nLimited genetic structure and high gene flow in Fasciola hepatica populations infecting 1 \nruminants in different geographic areas in the UK 2 \nMuhammad Abbas1, Kezia Kozel1, Nick Selemetas2, Olukayode Daramola3, Eric R. Morgan4, 3 \nUmer Chaudhry5, Martha Betson1* 4 \n1 Discipline of Comparative Biomedical Sciences, School of Veterinary Medicine, University 5 \nof Surrey, Guildford, UK 6 \n2 Discipline of Microbes, Infection and Immunity, School of Veterinary Biosciences, 7 \nUniversity of Surrey, Guildford, UK 8 \n3 School of Veterinary Medicine, University of Lancashire, UK 9 \n4 School of Biological Sciences, Queen's University, Belfast, UK 10 \n5Department of Veterinary Biomedical Sciences, Lewyt College of Veterinary Medicine, Long 11 \nIsland University, USA 12 \n* Corresponding author: Martha Betson, m.betson@surrey.ac.uk 13 \n 14 \n 15 \n 16 \n 17 \n 18 \n 19 \n 20 \n 21 \n 22 \n 23 \n 24 \n 25 \n 26 \n 27 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n2 \n \nAbstract 28 \nThe liver fluke, Fasciola hepatica, is a major parasitic threat to ruminant health and 29 \nproductivity worldwide, with important implications for food security, animal welfare, and 30 \nzoonotic risk. This study developed and validated a multiplex deep amplicon sequencing 31 \nassay targeting the mitochondrial NADH dehydrogenase 1 (mt-ND1) and cytochrome c 32 \noxidase subunit 1 (mt-COX1) loci for high-throughput genotyping of F. hepatica. DNA was 33 \nextracted from eggs sedimented from sheep and cattle faeces (n = 78) received from farms 34 \nand from adult worm pools (n = 12) isolated at abattoirs from diverse regions across the UK. 35 \nFollowing high-throughput sequencing, bioinformatics analysis was performed to 36 \ndemultiplex Illumina sequence reads and extract amplicon sequence variants (ASVs). A total 37 \nof 11 ASVs were identified at each locus (mt-ND1: 264–279 bp; mt-COX1: 312–319 bp), with 38 \ntwo or three predominant ASVs per locus, along with rare variants. Network and PCA 39 \nanalyses revealed two distinct clusters at the mt-ND1 locus: one primarily associated with 40 \nsheep and another shared between sheep and cattle. In contrast, mt-COX1 sequence reads 41 \nformed a single dominant cluster. Population analyses revealed extensive ASV sharing 42 \nacross regions, indicating high gene flow, likely facilitated by livestock movement and 43 \nparasite adaptation. 44 \nKeywords: F. hepatica, mt-ND1, mt-COX1, population genetics, ruminants, UK 45 \n 46 \n  47 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n3 \n \nIntroduction 48 \nThe liver fluke genus Fasciola comprises two species: Fasciola gigantica and Fasciola hepatica. 49 \nF. hepatica is prevalent in temperate zones , including the UK, Europe, parts of Oceania and 50 \nthe Americas [1], however, both species can be found in tropical and subtropical regions of 51 \nAsia and Africa, and their hybrids are also prevalent in Asia and Africa [1,2]. The lifecycle of 52 \nFasciola is complex, with various definitive mammalian hosts, including sheep and cattle. The 53 \nintermediate hosts in this lifecycle are mud snails: Galba truncatula [3], previously known as 54 \nLymnaea truncatula, for F. hepatica [4] and L. natalensis for F. gigantica [5]. This parasite is 55 \ntransmitted through food plants and herbage contaminated with  metacercariae, infecting 56 \nmainly small and large ruminants, though other mammalian species including humans can be 57 \ninfected [6]. 58 \nVarious factors can influence the occurrence of the parasite and resulting disease, including 59 \nenvironmental conditions such as rainfall [7], moisture levels, and temperature ( with 60 \ntemperatures from 10°C to 25°C  being optimal), as well as  the geography of grazing areas  61 \n(e.g., topography and soil type)  [8–11] and animal movement [12]. A feature of the lifecycle 62 \nis the clonal expansion of Fasciola spp. within its intermediate snail host, which contributes 63 \nto pasture contamination and to the subsequent infection of hosts by metacercariae of  the 64 \nsame genetic origin. This clonal expansion may lead to a genetic bottleneck effect  in the 65 \nparasite, particularly when infection levels in snail populations are low [13]. 66 \nUnderstanding the population genetics of F. hepatica infection provides crucial insights to 67 \naid the design of effective control strategies [14]. Over the past few decades, high levels of 68 \nanimal movement have been reported in domestic ruminants in several European countries 69 \n[15,16]; hence, analysing the population genetic structure of F. hepatica can assist in 70 \nunderstanding the corresponding spread of parasites infections. Determining parasite 71 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n4 \n \npopulation genetics can further inform on transmission dynamics and infection rates, 72 \nthereby helping identify interventions to reduce disease burden [12]. For example, if a 73 \nparasite population is characterised by a single dominant amplicon sequence variant (ASV) 74 \nat high frequency, this suggests a single infection source with clonal expansion of the 75 \nparasite in the intermediate host, with low metacercariae mixing in pasture settings [12,13]. 76 \nOn the other hand, multiple ASVs at varying frequencies in parasite populations might 77 \nindicate multiple infection sources on the farm and high mixing of metacercariae [13,17].  78 \nFew studies of large and diverse fluke populations examine whether infection has emerged 79 \nrecently in the host at a single time point, or whether burdens have been established 80 \nrepeatedly at different times before spreading. Recently, we have used these methods to 81 \nstudy the multiplicity of Calicophoron daubneyi infection in the United Kingdom [17] and 82 \nFasciola gigantica infection in Pakistan [12]. Our findings were consistent with multiple 83 \nindependent emergences of C. daubneyi infection, while the identification of common 84 \nvariants across several populations spanning a range of geographic locations highlights the 85 \nrole of animal movements in the parasite’s spread [17]. Moreover, our findings also suggest 86 \nthat most of the hosts were predominantly infected with the emergence of F. gigantica 87 \ninfection, while the identification of identical variant, consistent with clonal multiplication 88 \nwithin the snails. The most common variants was identified across several populations 89 \nspanning a range of geographic locations, again highlighting the role of animal movements 90 \nin the spread of F. gigantica infections [12]. 91 \nPopulation genetics can be determined using mitochondrial DNA (mtDNA) markers [18–22], 92 \nas well as nuclear microsatellite loci [23]. Microsatellites are usually highly polymorphic 1-6 93 \nbp sequences that can be used as markers to investigate genetic diversity and  genetic 94 \ndifferentiation using genomic DNA [24]. Recently, a panel of 15 highly polymorphic nuclear 95 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n5 \n \nmicrosatellite loci tested on DNA extracted from different life cycle stages, such as eggs, 96 \nadult worm, miracidia and metacercariae has been reported for F. hepatica [23]. High levels 97 \nof genetic diversity and clonal expansion of parasite and panmictic (randomly mating) 98 \npopulations has been reported by using microsatellite markers on an abattoir-based 99 \npopulation genetics study of F. hepatica in cattle in the central England and Wales [13]. 100 \nThere are several advantages of using microsatellite markers include their distribution over 101 \neukaryotic nuclear genomes [25], and mutation information can be useful for calculating 102 \nHardy-Weinberg Equilibrium (HWE) to study homozygosity and heterozygosity [23,24] 103 \nbecause genomic DNA contains information on inheritance from both parents [26]. 104 \nHowever, limitations include high mutation rates and elevated levels of polymorphism [27]. 105 \nFurthermore, a high number of alleles per locus in microsatellites can inflate F-statistic 106 \nvalues [28] and confound interpretation. Thus, this issue can sometimes lead to over- or 107 \nunderestimates of genetic diversity when the most common allele occurs at either very low 108 \nor very high frequencies [29]. Microsatellite datasets can be also prone to genotyping 109 \nerrors, which can bias downstream population genetic analyses [30]. 110 \nMitochondrial DNA is frequently used as a marker to study population genetics because of 111 \nits haploid nature [31], maternal inheritance [32], high copy number [33], clock-like and 112 \nneutral evolution rates [34], and a lack of recombination after heteroplasmy [35]. These 113 \ncharacteristics also make mitochondrial markers valuable for investigating the genetic 114 \ndiversity and genetic differentiation in fluke populations. Mitochondrial markers have been 115 \nwell documented for studying population genetics in Fasciola spp. from different regions of 116 \nthe world [12,14,22].  117 \nAs sequencing technology advances, deep amplicon sequencing enables the study of the 118 \npopulation genetics of Fasciola spp. in greater depth [12,21,23]. For population genetics 119 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n6 \n \nstudy,the mitochondrial marker mt-ND1 was used for F. gigantica and mt-COX1 for C. 120 \ndaubneyi, along with deep sequencing, utilising DNA isolated from adult worms of naturally 121 \ninfected cattle and sheep [17,36]. Deep sequencing enables the detection of dominant and 122 \nlow-frequency variants in different parasite populations [36], which can be useful to study 123 \ninsights into parasite transmission intensity, infection sources, and metacercarie mixing. 124 \nHowever, assays for assessing genetic diversity of F. hepatica using deep sequencing of 125 \nmultiplexed mitochondrial markers yet not exist. Such assays would provide opportunities 126 \nto shed new light on the population genetics of F. hepatica from naturally infected hosts 127 \nacross various regions of the UK. For example, the extent of overlap between sheep and 128 \ncattle, and the presence of geographical clustering, could provide insights into the roles of 129 \npasture sharing and livestock movement in driving fluke infection and disease. 130 \nThe present study aimed to develop and test  an improved deep amplicon sequencing 131 \napproach to investigate the population genetics of F. hepatica infections across the UK using 132 \ntwo mitochondrial markers (mt-ND1 and mt-COX1). This multiplexed sequencing method and 133 \nhigh-throughput sequencing approach reduces experimental complexity for studying host-134 \nlevel population genetic s in F. hepatica populations using a single Illumina sequencing run.  135 \nThis methodology enabled the examination of transmission dynamics and gene flow in both 136 \nadult worm and egg DNA obtained from natural F. hepatica infections in the UK.  137 \nResults 138 \nValidation of multiplex PCR and demultiplexing of mitochondrial markers  139 \nThe multiplex meta-barcoded PCR targeting mitochondrial positive control F. hepatica DNA 140 \nsuccessfully amplified distinct bands at 311 bp for mt-ND1 and 359 bp for mt-COX1, with no 141 \nevidence of non-specific amplification (Supplementary Fig. 1a, b and c). 142 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n7 \n \nOut of a total of 90 individual samples processed, 78 (86.66%) samples generated sequence 143 \nreads for the mt-ND1 marker (n = 12 adult worm DNA, n = 66 egg DNA). These sequences 144 \nobtained from mt-ND1 loci were categorised into 40 parasite populations, comprising cattle 145 \n(n = 14) and sheep (n = 26) populations. For the mt-COX1 marker, 84 (93.33%) samples 146 \nsuccessfully produced sequence reads (n = 12 adult worms, n = 72 egg DNA), grouped into 147 \n42 parasite populations, including cattle (n = 15) and sheep (n = 27). 148 \nGeographical distribution of mt-ND1 locus ASVs of F. hepatica 149 \nEleven ASVs were identified at the F. hepatica mt-ND1 locus (accession numbers PX902280-150 \nPX902290), ranging from 264 bp to 270 bp, using a library of mt-ND1 reference sequences 151 \ndownloaded from NCBI database (Supplementary Fig. 2), and their frequencies were 152 \nrecorded across 40 fluke populations in different counties of the UK (Fig. 1a and 153 \nSupplementary Table 1). Across the 40 parasite populations analysed, a total of 1,403,462 154 \nsequence reads were extracted, of which the majority (1,167,674 reads; 83.1%) belonged to 155 \na small number of predominant ASVs, including ASV1, ASV2, and ASV3. These predominant 156 \nASVs reflect the dominant variants circulating in cattle and sheep in different regions. 157 \nASV1 was the most abundant variant (39% of total reads), detected in 18 populations across 158 \n10 counties (Supplementary Table 2). It was predominant in 16 populations across seven 159 \nregions, exceeding 95% dominance in sheep-derived populations in Southern Scotland and 160 \nthe West of Scotland (Supplementary Table 1). 161 \nASV2 was the second most abundant variant (35.9% of total reads), found in 24 populations 162 \nacross 13 counties (Supplementary Table 2) covering nine regions. It was predominant in 15 163 \npopulations, reaching greater than 99% in West Midlands England and South-East England 164 \nsheep flocks (Supplementary Table 1). ASV3 ranked third (14.7% of total reads), occurring in 165 \n9 populations in eight counties (Supplementary Table 2) and in seven populations, being the 166 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n8 \n \nonly ASV present  in cattle and sheep fluke in North West England, as well as in sheep fluke 167 \nin South West England and the East Midlands (Supplementary Table 1). 168 \nIn South-East and North West England, ASV3 and ASV1 predominated in fluke populations in 169 \ncattle and in South East England in fluke populations in sheep ASV2 predominated, with 170 \nminor contributions from ASV8. East of England stands out for about equal proportions of 171 \nASV1 and ASV3 suggesting within-farm genetic mixing. (Supplementary Table 1). 172 \nIn the Scottish Borders, ASV1 and ASV2 predominated in cattle fluke with ASV6 a rare 173 \nvariant and sheep fluke populations had ASV1 (68.66%) and ASV4 (28.98%) as the major 174 \ncontributors, and ASV2 and ASV8 as minor contributors. Southeastern Scotland cattle fluke 175 \nexhibited ASV3 dominance (89.19%), with minor proportions of ASV1 and ASV2. In the West 176 \nof Scotland, most sheep fluke populations ASV2 was most common and one population had 177 \nASV10 (39.52%), a variant otherwise absent from the other populations. In Northern 178 \nIreland's County Tyrone, fluke in a single sheep population predominantly showed ASV2 179 \n(78.16%), and other varients were ASV9 (18.15%) and ASV8 (3.69%), (Fig. 1a, Supplementary 180 \nTable 1). 181 \nNotably, some ASVs were found in some specific regions and fluke populations. For 182 \nexample, ASV4 was found locally in two Southern Scotland sheep populations (P21S, 77.62% 183 \n(dominant); P25S, <0.1% (rare)). Similarly, ASV6 was rare but reached 16.1% in a single 184 \nScottish Borders cattle population (P14C). ASV7 reached a high frequency (25.41%) in a 185 \nsingle Southern Scotland sheep flock (P20S). ASV8 was a widely distributed but rare variant 186 \nobserved at ≤3% abundance but present across multiple regions. Moreover, ASV9 was 187 \nprimarily found in Northern Ireland, while ASV10 was detected in the West of Scotland 188 \nsheep flock, indicating geographically isolated variants (Supplementary Table 1). 189 \nNetwork trees clustering analysis of the mt-ND-1 locus 190 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n9 \n \nThe Neighbour-Net network tree with pie charts confirmed a highly connected parasite 191 \npopulations dominated by two central, multi-regional ASVs, including ASV1 (the largest 192 \nnode) and ASV2 (the second-largest node), which were present in both cattle and sheep 193 \nfluke across multiple regions (Fig. 2a and b). ASV1 and ASV2 were connected via ASV3 or 194 \nlow-abundance ASV8, a genetic link between England and Scotland cattle fluke-derived 195 \ndominant ASVs and Scotland sheep fluke-derived dominant ASVs. Peripheral ASVs for ASV1 196 \nincluded ASV5, ASV6, ASV9, ASV10, and ASV11. Peripheral ASVs for ASV2 were ASV4 and 197 \nASV7. Notably, all 11 ASVs were found in sheep fluke. In comparison, four ASVs (ASV10, 198 \nASV11, ASV5, and ASV7) were not detected in cattle fluke (Fig. 2b). Median Joining Network 199 \ntree of mt-ND1 (Supplementary Fig. 3a) showed similar linkages among ASVs to the 200 \nNeighbour-Net tree (Fig. 2).  201 \nThe mt-ND1 PCA plot based on sequence reads of 11 ASVs from 40 populations showed 202 \npartial clustering of F. hepatica populations by host and region, with the first two principal 203 \ncomponents explaining PC1 (18.31%) and PC2 (15.18%) of the total sequence read data 204 \n(33.34%) (Fig. 3a). Populations of F. hepatica in sheep and cattle across the UK showed both 205 \nsubstantial overlap and some degree of regional clustering, indicating high gene flow with 206 \noccasional location-specific patterns. Most sheep derived populations fell into Cluster 1, 207 \nfrom 9 geographical regions including South East England, East Midlands England, South 208 \nWest England, North West England, West Midlands England, Scottish Borders, Southern 209 \nScotland, West of Scotland, and Northern Ireland. There were two regions not found in 210 \ncluster 1 including East of England, and Southeastern Scotland. This suggests that sheep 211 \nacross different areas carry genetically similar parasite variants. Sheep fluke populations are 212 \nmore evenly distributed between Cluster 1 and Cluster 2. In Cluster 2, the parasite 213 \npopulation in sheep and cattle was widely distributed in 9 regions however, samples from 214 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n10 \n \nWest Midlands England and Northern Ireland were not represented in this cluster. This 215 \nindicates that the parasite variants are not geographically restricted and are prevalent in 216 \nboth host species. 217 \nA few populations demonstrated regional genetic divergence. For example, one sheep fluke 218 \npopulation from Southern Scotland (P21S) clustered apart with very high PC1 and PC2 219 \nvalues, suggesting unique ASVs in the area. Similarly, Scottish Borders sheep (P9S) and 220 \nSouthern Scotland sheep (P20S) also showed genetic separation from the main clusters.  221 \nThe split topology tree of mt-ND1 sequences showed that one cluster is dominated by ASV1 222 \nand groups closely with ASVs 9 and 10. The second cluster is defined by ASV2 and ASV3, 223 \nindicating an evolutionary relationship between these variants. Remaining ASVs (ASV4–224 \nASV11) are distributed along shorter branches in between main variants, representing low-225 \nfrequency variants that are genetically closer to one of the two dominant clusters (Fig. 3b). 226 \nGenetic diversity analysis of the mt-ND-1 locus 227 \nThe analysis of molecular variance (AMOVA) from PopArt revealed that the majority of 228 \ngenetic variation in F. hepatica mt-ND1 populations occurred within populations (137.85%). 229 \nVariation among groups accounted for only 1.31% of the total, and variation among 230 \npopulations was negative (–39.17%) (Table 1, Supplementary Fig. 3a). The negative variance 231 \ndetected in the AMOVA results showed lack of genetic differentiation and should be 232 \ninterpreted as zero [37,38]. All fixation indices were low and non-significant (Phi ST = –233 \n0.3785, P = 1.000; Phi SC = –0.3969, P = 0.997; Phi CT = 0.0131, P = 0.544), confirming the 234 \nabsence of significant genetic differentiation between regions or populations. Similar 235 \nAMOVA results were confirmed by Arlequin (Supplementary File 1). These values suggest 236 \nhigh genetic connectivity and gene flow among populations across the UK, consistent with 237 \nthe sharing of common ASVs between regions and hosts. Overall, nucleotide diversity 238 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n11 \n \n(π=0.00502) indicated low genetic variation in the population. Tajima's D neutrality test was 239 \nnot significant -0.372, (p = 0.623). 240 \nGeographical distribution of mt-COX1 locus ASVs of F. hepatica 241 \nA total of 11 ASVs (312 bp to 319 bp) (accession numbers PX861700-PX861710) were 242 \nidentified at the F. hepatica mt-COX1 locus, using reference sequences of mt-COX1 243 \n(Supplementary Fig. 4), and their frequencies were recorded in 42 populations in different 244 \ncounties of the UK (Fig. 1b, and Supplementary Table 3). A total of 1.764 million sequence 245 \nreads were generated across the 42 parasite populations, of which 1.33 million reads 246 \n(75.2%) were from two predominant ASVs and 437,229 (24.8%) reads corresponded to rare 247 \nASVs (Supplementary Table 3). ASV1 and ASV2 were the dominant variants circulating in 248 \ncattle and sheep fluke in different regions. The rare ASV3 to ASV11 were often 249 \ngeographically restricted. 250 \nThe most abundant variant was ASV1, contributing 45.0% of sequence reads overall and 251 \ndetected in 41 populations across all 17 counties (Supplementary Table 4). It was the 252 \npredominant variant found both in cattle and sheep fluke in 22 populations, frequently 253 \nrepresenting 75% of sequence reads. For example, in sheep fluke, ASV1 was observed in the 254 \nWest Midlands England (100% of reads), South East England (>99%), North West England 255 \n(96%), Southern Scotland (99.46%), in the West of Scotland (>70%), Scottish Borders 256 \n(86.87%) and a major variant in Northern Ireland (68.23%). In cattle, ASV1 was common in 257 \nSouth West England (88.81%) and the Scottish Borders (75.2%). In North West England, 258 \nASV1 and ASV2 were found in equal proportions of 50% each (Supplementary Table 3). 259 \nASV2 ranked second in abundance, with overall sequence reads of 45.1% in 38 populations 260 \nand 16 counties. ASV2 was found to be dominant in cattle and sheep fluke across 18 261 \npopulations from 8 regions. It was widespread in cattle fluke from the North West (>95%), 262 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n12 \n \nSouth West (>66.34%), and South East England (86.67%). ASV2 was highly prevalent in 263 \nsheep fluke across regions of the UK including North West England (81.07%), East of England 264 \n(75.77%), the West of Scotland (81.69%) and Southern Scotland (>52.06%) (Supplementary 265 \nTable 3). Of the other variants only ASV6 and ASV10 were common. ASV6 predominated in 266 \nEast Midlands sheep fluke (90.95%), and ASV10 was predominant in Scottish Borders cattle 267 \nfluke (87.5%). The remaining ASVs were present only as rare variants in different 268 \npopulations (Supplementary Table 3). 269 \nAlthough ASV1 and ASV2 were found to be predominant, the analysis of rare variants 270 \nhighlighted their presence in many regions, including Northwest England, Southwest 271 \nEngland, Southern Scotland, and South Lanarkshire, ranging from  0.04% to 46.5% of 272 \nsequence reads. There were certain rare variants noted in sheep and cattle fluke, for 273 \nexample, in sheep fluke ASV3 (24.2%) was in the East of England, ASV4 (34.01%) and ASV5 274 \n(12.64%) appeared in Scottish Borders, and ASV7 (<0.01%) in East of England. In cattle, ASV3 275 \n(0.04%) in South East England ASV5 (1.31%) in South West England, ASV7 (5.51%) occurred 276 \nin Scottish Borders, and ASV8 was noted 13.29% and <0.01% in South East England and 277 \nScottish Borders, respectively. There were rare variants found in both hosts including ASV9 278 \n(< 4%) in South West England, East Midlands England, Scottish Borders, South Lanarkshire, 279 \nand West of Scotland. ASV10 (0.53% to 29.3%) appeared in Scottish Borders, South Eastern 280 \nScotland, and in South West England, while ASV11 (<0.01% to 2.08%) in North West 281 \nEngland, Southern Scotland, Northern Ireland, South East England, and East Midlands 282 \n(Supplementary Table 3). Overall, the rare ASVs demonstrated regional and host-specific 283 \npatterns and may emerge in the future in both sheep and cattle populations across the UK. 284 \nNetwork trees clustering analysis of the mt-COX1 locus 285 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n13 \n \nThe Neighbour Net analysis of the 11 mt-COX1 ASVs revealed two main nodes. ASV1 and 286 \nASV2  were connected through the low-abundance ASVs 4, 6, 10, and 11, forming a genetic 287 \nbridge between host- and region-specific ASVs (Fig. 4a). The peripheral ASVs for ASV1 288 \nincluded ASV3, ASV7, ASV8, and ASV9. For ASV2, there was only ASV5.  289 \nFor instance, ASV1 was most abundant in sheep fluke across Scotland and England, but was 290 \nalso found in considerable amounts in cattle across the same regions. ASV2 was detected in 291 \nsheep and cattle fluke in England, as well as in sheep fluke in Scotland, with low abundance 292 \nin sheep fluke from Northern Ireland and in cattle fluke from Scotland. In contrast, other 293 \nASVs showed strong geographic and host specificity. ASV3 and ASV6 were found in English 294 \nsheep, ASV4, ASV5, and ASV9 were found in Scottish sheep, ASV7 was specific to Scottish 295 \ncattle, ASV8 was restricted to English cattle, and ASV11 was found mainly in Northern 296 \nIreland sheep (Fig. 4b). 297 \nMedian Joining Network tree of mt-COX1 (Supplementary Fig. 3b) showed similar linkages 298 \namong ASVs as the Neighbour Net tree (Fig. 4a and b). 299 \nThe PCA plot based on mt-COX1 sequence reads from 42 F. hepatica populations showed a 300 \nsingle cluster by host and region, with the first two principal components explaining PC1 301 \n19.07% (PC1) and 15.69%(PC2) of the total (34.76%) (Fig. 5a). All parasite populations in 302 \nsheep and cattle fluke across the UK show substantial overlap and indicating high gene flow. 303 \nThe topology tree of 11 mt-COX1 ASVs showed two major ASVs (ASV1 and ASV2) at the ends 304 \nof the tree, and the remaining ASVs were mainly distributed along shorter branches in 305 \nbetween the main ASVs, indicating phylogenetic separation among ASVs (Fig. 5b).   306 \nGenetic diversity analysis of the mt-COX1 locus 307 \nThe AMOVA results showed that the genetic variation occurred within populations (139.4%), 308 \nwhile variation among groups (0.75%) and among populations within groups (−40.2%) was 309 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n14 \n \nvery low (Table 2, Supplementary Fig. 3b). Similar AMOVA results were found using Arlequin 310 \n(Supplementary File 2). Consistently, fixation indices were very low or negative (Phi ST = 311 \n−0.394, Phi SC = −0.405, Phi CT = 0.007), and none were statistically significant (p > 0.05). 312 \nTogether, these results suggest that there is no significant genetic variation among groups 313 \nand populations. Genetic diversity was found only within populations, not across hosts or 314 \ngeographic regions. Overall, nucleotide diversity (π = 0.00834) was low across all 315 \npopulations. Tajima's D test was 0.375 and not significant (p = 0.343). 316 \nDiscussion 317 \nThe present study developed a metabarcoding approach using multiplex mitochondrial 318 \nmarkers targeting the mt-ND1 and mt-COX1 loci to study the population genetics of F. 319 \nhepatica, using samples from natural infections collected across different areas of the UK. 320 \nThe application of the metabarcoded multiplexed markers in deep amplicon sequencing can 321 \nprovide an efficient tool for investigating genetic diversity patterns in multiple populations 322 \nof F. hepatica that can further inform transmission dynamics and infection rates. This study 323 \ndemonstrated that F. hepatica populations with a small number of predominant ASVs were 324 \ncirculating in both cattle and sheep with high gene flow across different regions of the UK. 325 \nAlongside this a few rare geographically restricted ASVs were also noted. The results 326 \nhighlighted a pattern of widespread ASV flow across the UK, which may be driven by clonal 327 \npropagation of the parasite in intermediate snail hosts, livestock movement, grazing 328 \npractices, and parasite adaptation to the UK environment.  329 \nThis study utilised mitochondrial markers to analyse the genetic diversity of F. hepatica 330 \npopulations because mitochondrial DNA is useful for population genetics investigations, due 331 \nto its high copy number [33], maternal inheritance [32], and transmission without 332 \nrecombination under neutral heteroplasmy conditions [35]. These properties make 333 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n15 \n \nmitochondrial DNA a powerful target for investigating evolutionary studies. Mitochondrial 334 \nmarkers have been used to study genetic diversity and infection dynamics in F. gigantica, 335 \ntargeting the mt-ND1 locus where a single mt-ND1 variant of F. gigantica was predominant 336 \nin most of the hosts in Pakistan [12], and to investigate C. daubneyi, using the mt-COX1 337 \nlocus. Notably, multiple variants of C. daubneyi infections, were detected across different 338 \ngeographic regions of the UK [17]. A study from Malawi supported the suitability of both mt-339 \nND1 and mt-COX1 loci for population genetics studies in F. gigantica and reported recent 340 \npopulation expansion [22]. Other studies also supported the use of mitochondrial markers 341 \nfor genetic diversity and gene flow studies in F. hepatica [39–42]. 342 \nThe amplification success across 90 samples, with 78 and 84 yielding sequence reads 343 \nfor mt-ND1 and mt-COX1, respectively, highlighted the effectiveness of our method. 344 \nFurther, the deep amplicon sequencing technique detected both dominant and rare ASVs, 345 \nenhancing our understanding of parasite population genetics. This ability to recover ASVs in 346 \ndifferent parasite populations across geographical regions of the UK demonstrated that 347 \nmultiplex PCR combined with next-generation sequencing can be a valuable tool for 348 \nstudying fine-scale genetic structure. Previous studies used PCR-RFLP, conventional PCR, 349 \nand Sanger sequencing to amplify and analyse the mitochondrial genome regions to 350 \ninvestigate genetic variations [22,43,44]. However, these methods are low-throughput, 351 \ntime-consuming, relatively expensive when handling medium to high numbers of samples 352 \n[45]. In contrast, high-throughput deep amplicon sequencing using the Illumina MiSeq 353 \nplatform offers a convenient and cost-effective method [46] for handling a medium to high 354 \nnumber of samples, with thousands and millions of sequence reads generated per 355 \npopulation in a single run [12,17,47]. 356 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n16 \n \nThis study identified some predominant ASVs across different regions of the UK, as well as 357 \nmore locally restricted variants. For instance, the most common three variants of mt-ND1 358 \n(ASV1-3) were widely distributed across multiple regions and hosts and together accounted 359 \nfor over 80% of infections. Similarly, two mt-COX1 variants were dominant. AMOVA results 360 \nshowed that most genetic variation occurred within populations rather than between 361 \ncounties or regions, which is consistent with high level of gene flow. Based on microsatellite 362 \ngenotyping, previous research in the UK also found high genetic diversity and gene flow, as 363 \nwell as an absence of defined population structures. This was believed to be due to the 364 \nclonal emergence of F. hepatica infections through the intermediate snail host [13], such 365 \nthat a single miracidium infecting a G. truncatula mud snail can generate multiple 366 \ngenetically identical cercariae [13]. However, metacercariae may mix on pasture, resulting in 367 \na more varied genetic profile before ingestion by the definitive host [12].    368 \nFindings from F. hepatica isolates from three geographical regions of China supported our 369 \nresults, showing that genetic variation can occur within populations rather than between 370 \npopulations [48]. In another example from the European context, in the Netherlands, 371 \ngenetic diversity was also reported mostly within populations rather than between 372 \npopulations [49]. In Algeria, low genetic diversity and a common origin for the parasite's 373 \ncountrywide distribution were reported, with only two variants from the mt-COX1 gene 374 \n[50]. In Colombia, no genetic diversity was found among F. hepatica parasites. However, the 375 \nauthors mentioned that this might be due to the low resolution of the molecular markers 376 \nused including nuclear markers (28S, β-tubulin 3, ITS1, ITS2), and mitochondrial marker (mt-377 \nCOX1) [51]. 378 \nIn contrast, high genetic diversity between populations was found in German dairy cattle 379 \nusing mitochondrial (mt-ND1 and mt-COX1) and eight microsatellite markers [44], in cattle 380 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n17 \n \nand sheep in Spain and Peru with mt-ND1 marker [52], and in grazing cattle in Australia 381 \nusing mt-ND1 and mt-COX1 markers [42]. Further, a high number of mt-COX1 mitochondrial 382 \nvariants have been reported in cattle and horses in Chile [53]. A 2007 study reported that a 383 \nsingle animal in Ireland could harbour ten distinct mitochondrial variants of F. hepatica, and 384 \nthe author related that this genetic diversity predates the last ice age [43]. A global-scale 385 \nanalysis of NCBI data showed that both mt-ND1 and mt-COX1 locus-specific variants were 386 \ncirculating in different parts of the world, with high Tajima's D values and a low likelihood of 387 \nfuture population growth [54]. However, from Armenia, Algeria, Brazil, Spain, and Ecuador, 388 \nnegatively significant Tajima's D values were reported for mt-ND1 along with mt-COX1 389 \nshowed deviation from neutrality, supporting recent population expansion [54]. This 390 \ncontrast between the neutral, locally mixed parasite populations found in this work from UK 391 \nand the variable patterns observed globally highlights how local gene flow can influence in 392 \nfuture evolutionary processes in F. hepatica populations. 393 \nUnlike the predominant ASVs, the rare ASVs identified in this study exhibited a mostly regional 394 \ndistribution, with only a few rare ASVs found in multiple populations across different areas, 395 \nfor example, mt -ND1 (ASV8) and mt-COX1 (ASV9-ASV11). Th is showed the emergence of 396 \nlocalised variants , potentially linked to specific environmental factors, especially 397 \nenvironmental temperature and soil conditions, which can influence th e transmission 398 \ndynamics of F. hepatica [55]. For example, G. truncatula egg masses are rarely observed in 399 \nshaded areas  [56], and this could affect opportunities for fluke transmission in different 400 \nenvironments. Fasciola eDNA was less frequently detected in dark brown or black soils, which 401 \nare rich in organic matter in the form of peat and have lower pH values [56]. This could impact 402 \nthe distribution of intermediate snail hosts and the  clonal expansion of rarer ASVs . Further, 403 \nFasciola spp. egg hatching and development are optimal between 20 °C and 30 °C and 404 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n18 \n \ninhibited at temperatures below 10 °C. The time required for miracidia hatching decreased 405 \nwith increasing temperature, and shedding of cercariae from snail hosts was most rapid  at 406 \naround 27 °C  [55]. Viability of metacercariae declined at higher temperatures but could be 407 \nprolonged under high humidity. Snails grow best at 25 °C, and their susceptibility to Fasciola 408 \ninfection is also temperature dependent [55]. Climate variation and change could therefore 409 \nimpact the geographic distribution of liver fluke, and drive adaptation to local conditions. For 410 \ninstance, F. hepatica infections have historically been low in regions such as Southern Europe, 411 \nbut climate shifts may increa se winter risk due to temperature and moisture fluctuations  412 \nfalling into suitable ranges [11]. Increased fluke infections have been recorded in the EU and 413 \nthe Northern Altiplano in South America [57]. Additionally, a study  in New Zealand utilised 414 \nhistoric climate data  (1972-2012) and predicted that areas with low initial risk, such as 415 \nCanterbury and Otago, could see a near 200% rise by 2090 [58]. Although we have currently 416 \nidentified rare and locally restricted ASVs, these studies highlight how they could serve as 417 \npotential reservoirs of  future genetic diversity that could become epidemiologically 418 \nsignificant under shifting environmental conditions. Further research is needed to determine 419 \nhow genetic diversity measured using different markers to differences in biological responses 420 \nand environmental conditions by F. hepatica and other parasites. 421 \nIn our study, Neighbour-Net, median joining networks and PCA showed that UK F. hepatica 422 \npopulations are interconnected and dominated by a few ASVs. In the PCA plot, mt-ND1 423 \nrevealed one cluster containing mainly sheep-hosted parasite populations and a second 424 \ncluster containing both sheep and cattle-hosted populations. In contrast, for mt-COX1, 425 \nparasite populations from both sheep and cattle across different regions of the UK formed a 426 \nsingle, overlapping cluster. Our study did not gather data on co-grazing of sheep and cattle. 427 \nHowever, sheep and cattle often co-graze in Northern Ireland and may be infected with the 428 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n19 \n \nclonal variants of parasites [43]. Prichard et al., 2005 also attributed fluke outbreaks in East 429 \nEngland to co-grazing with sheep imported from other areas and high rainfall during 430 \nsummer [59]. This homogenisation of ASVs across populations in England, Scotland, and 431 \nNorthern Ireland is likely due to livestock movement, a common practice in the UK 432 \nagricultural system and important for the spread of various diseases [60,61]. The role of 433 \nanimal movement in parasite transmission and high gene flow has been well documented 434 \n[12,13,17,62]. Furthermore, greater genetic structure was observed in parasite populations 435 \nin sheep rather than cattle, which may be due to differences in grazing behaviour. Sheep 436 \ntend to feed closer to soil and waterlogged areas than cattle, potentially increasing 437 \nexposure [63]. In contrast, higher genetic diversity was reported in cattle fluke than in sheep 438 \nand goat fluke in Iran [64], but this may be because the prevalence of this disease is higher 439 \nin cattle than in sheep in Iran [65,66]  440 \nThe present study has limitations. The use of only two mitochondrial markers (mt-ND1 and 441 \nmt-COX1) provided valuable resolution but captures only maternally inherited variation. The 442 \nuse of nuclear markers or whole-genome data could further enhance understanding of 443 \nnuclear-level population genetics, and host adaptation [13,23,67]. Although 90 samples 444 \nfrom 17 counties were analysed, the sample coverage was uneven, with some regions 445 \nrepresented by only one or two populations, and no faecal samples were obtained from 446 \nNorthern Ireland. Unequal sampling can overrepresent genetic structuring and overstate the 447 \napparent dominance of a few ASVs in different regions. Finally, intermediate host snail and 448 \nlivestock movement data were not included in this study, although both are known to 449 \ninfluence the spread of F. hepatica infection strongly. 450 \nFuture research should aim to overcome the stated limitations by optimising nuclear 451 \nmarkers and whole-genome sequencing to complement mitochondrial markers and capture 452 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n20 \n \ngenetic polymorphisms and adaptive behaviours (manuscript in preparation). A higher 453 \nnumber of samples across multiple years and seasons would help to track deeper dynamics 454 \nof ASV diversity. Linking parasite genetic data with phenotypic outcomes, such as flukicide 455 \nefficacy through faecal egg count reduction tests, infection intensity, and productivity losses 456 \nin cattle and sheep, will strengthen knowledge of fluke epidemiology and control options. 457 \nAssessment of genetic diversity of F. hepatica in intermediate snail hosts will provide 458 \ninsights into bottlenecks and parasite persistence in wetlands, while combining parasite 459 \ngenetics with livestock trade, and grazing movement data will enable causal inference about 460 \nhow gene flow is maintained across the UK. These approaches can generate more 461 \nunderstanding of F. hepatica transmission and evolution, supporting targeted interventions 462 \nagainst fasciolosis. The methods optimised and described here provide an additional tool for 463 \ncollecting genetic data and linking it with infection and disease outcomes, as well as 464 \ninferring patterns of parasite transmission and spread. 465 \nConclusion 466 \nIn conclusion, a multiplex mitochondrial metabarcoding approach has been developed here, 467 \nproviding a platform for medium to large-scale population genetic studies of F. hepatica 468 \ninfection. This study demonstrated that F. hepatica populations in the UK are largely 469 \ngenetically interconnected and dominated by a small number of widespread variants in both 470 \ncattle and sheep. The findings confirm that cross-transmission of fluke between co-grazing 471 \nsheep and cattle is likely, although some genotypes seem to be more restricted to sheep 472 \nfluke. In addition, rare and region-specific variants were found at low frequencies, which 473 \nmay contribute to the future emergence of new variants in the UK if not controlled. Our 474 \nfindings support the idea that high gene flow can result from parasite adaptation in the UK 475 \nenvironment alongside high levels of livestock movement. These findings are important for 476 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n21 \n \nunderstanding transmission dynamics, detecting emerging variants, and informing effective 477 \ncontrol strategies for F. hepatica infections to livestock farmers. 478 \nMethods 479 \nField samples 480 \nA total of 90 field samples comprising 78 faecal egg samples and 12 adult worm samples 481 \nwere selected, which identified as F. hepatica positive in our previous study [68]. Multiple 482 \nsamples collected from the same sheep and cattle farms, veterinary practitioners, or 483 \ncounties within close timeframes were merged into single parasite populations. 484 \nAdditionally, adult fluke obtained from abattoirs and at post-mortem examination from the 485 \nsame animal were treated as a single population. All populations were assigned by host 486 \nspecies, cattle (n=15) and sheep (n=27) for downstream analyses. 487 \nThe samples were collected across 17 counties in the UK between December 2022 and May 488 \n2024 in collaboration with cattle and sheep farmers, as well as registered veterinary 489 \npractitioners, in accordance with ethical approval NASPA-2122-04 [68]. The populations 490 \nwere further categorised into 11 regions across the UK including North West England 491 \n(Cheshire, Cumbria), East Midlands England (Derbyshire), West Midlands England 492 \n(Staffordshire), East of England (Essex), South East England (East Sussex, Kent, West Sussex), 493 \nSouth West England (Dorset, Devon, Gloucestershire, Wiltshire), Scottish Borders 494 \n(Peeblesshire), Southern Scotland (South Lanarkshire), Southeastern Scotland (West 495 \nLothian), West of Scotland (Renfrewshire), Northern Ireland (County Tyrone).  496 \nAdult worm populations were obtained from the livers of infected animals at abattoirs 497 \ndescribed [68]. All samples were transported to the School of Veterinary Medicine at the 498 \nUniversity of Surrey and stored at –20°C for subsequent analysis. 499 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n22 \n \nDNA extraction 500 \nDNA was extracted from a pooled sample of head tissue obtained from all available adult 501 \nworms per host and from faecal egg samples , following the methods described [68]. Elution 502 \nwas performed in 50 μL of molecular biology-grade water (Cytiva HyClone™), and the eluate 503 \nwas stored at –80 °C for subsequent analyses. 504 \nDevelopment of multiplex mitochondrial markers 505 \nPCR was conducted using Fasciola genus-specific mt-ND1 primers, resulting in a product size 506 \nof 311 bp [12]. In addition, mt-COX1 primers for F. hepatica were designed and tested in this 507 \nwork (Supplementary Table 5), potentially aiming for a product length of 319-475 bp. These 508 \nprimers were designed by aligning 435 mt-COX1 sequences from F. hepatica available on 509 \nGenbank and selecting a region showing variations and a conserved region after visualising 510 \ndifferent sequences using Primer3 in Geneious Prime version 8.0.5. The primer pair was 20 511 \nbp lengths, Tm values range was 59.2-59.8 °C, GC content 50–60%, and a forward and 512 \nreverse Tm difference of 0.6 °C. Primers did not contain any hairpins, self-dimers, and cross-513 \ndimers. 514 \nPCR conditions were first optimised using positive control DNA in a gradient PCR with 515 \nannealing temperatures ranging from 52°C to 60°C for amplification of mt-COX1 516 \n(Supplementary Fig. 1c). PCR was performed in duplicate and twice with 2 μl of the DNA 517 \ntemplate using DreamTaq Green PCR master mix (Thermo Scientific, USA) in a 25 μl reaction 518 \nmix and primer concentrations of 200 nM in the final volume. Final PCR conditions included 519 \n35 cycles of initial denaturation at 95°C for 5 minutes, followed by denaturation at 95°C for 520 \n1 minute, annealing of mt-ND1 primers at 50°C for 1 minute and mt-COX1 primers at 55°C 521 \nfor 1 minute, and extension at 72°C for 1 minute, with a final extension at 72°C for 5 522 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n23 \n \nminutes. Water was used as a no-template control. The resulting PCR products were 523 \nprocessed for Sanger sequencing. 524 \nA multiplex PCR was developed using both mt-ND1 and the mt-COX1 primers, with 525 \nDreamTaq Green PCR master mix (Thermo Scientific, USA) with primer concentrations of 526 \n200 nM in 25 μl final reaction volume. The PCR was conducted for 35 cycles with conditions 527 \nas follows: an initial denaturation step at 95°C for 5 minutes, followed by denaturation at 528 \n95°C for 1 minute, annealing at 53°C for 1 minute, extension at 72°C for 1 minute and a final 529 \nextension at 72°C for 5 minutes. The multiplex PCR reaction was carried out in duplicate 530 \nusing two μl of positive control F. hepatica DNA, and water as a negative control.  531 \nDeep amplicon sequencing of multiplexed mt-ND1 and mt-COX1  532 \nThe metabarcoded mt-ND1 [12] and mt-COX1 (Supplementary Table 1) markers were used 533 \nto target the mitochondrial DNA of F. hepatica. A multiplexed first-round PCR was carried 534 \nout using the KAPA HiFi PCR Kit (KAPA Biosystems, South Africa) as described [12]. The 535 \nsecond-round primer sets, adaptors, barcoded PCR amplifications, magnetic bead 536 \npurification, and final library quantification were based on previously described methods 537 \n[12,17,69]. 538 \nDemultiplexing of mt-ND1 and mt-COX1 sequences and bioinformatics analysis 539 \nThe Illumina MiSeq system demultiplexed the sequencing data based on sample-specific 540 \nbarcoded indices (Supplementary Table 6), generating corresponding FASTQ files for each 541 \nsample (NCBI Bioproject: PRJNA1402908, accession No: SAMN54606237-SAMN54606325, 542 \nhttps://data.mendeley.com/datasets/822rxwph9t/1). The resulting FASTQ files were further 543 \nanalysed using Mothur versions 1.41.0 and 1.48.1 [70] on the University of Surrey High-544 \nPerformance Computing (HPC) cluster. Sequence analysis was performed following the 545 \npipelines described in previous studies [12,17] with modifications as described below. 546 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n24 \n \nFor reference database construction, mt-ND1 and mt-COX1 sequences were retrieved from 547 \nthe NCBI database for F. hepatica and F. gigantica. A total of 363 mt-ND1 sequences and 548 \n462 mt-COX1 sequences were downloaded for F. hepatica, which were collapsed into 79 549 \nand 97 unique reference sequences, respectively. For F. gigantica, 351 mt-ND1 and 337 mt-550 \nCOX1 sequences were downloaded, resulting in 117 and 108 unique collapsing sequences, 551 \nrespectively. These reference sequence libraries for the mt-ND1 and mt-COX1 genes were 552 \nused for sequence demultiplexing and alignment. The Mothur pipeline joined paired-end 553 \nreads, filtered out ambiguous or low-quality sequences, and removed excessively long or 554 \nshort sequences. Sequences were aligned against the reference library, unique sequences 555 \nwere pre-clustered and abundant reads were then grouped. 556 \nASVs were obtained from the filtered dataset using minimum read thresholds assigned to 557 \neliminate noise and sequencing artefacts using the command \"split.abund\". A cutoff value 558 \nof 7,000 reads per ASV was applied to the mt-ND1 dataset, and 2,000 reads per ASV were 559 \nused for mt-COX1. These thresholds were determined empirically following visual inspection 560 \nof the output count table files. The aligned unique sequences were split into a high- and 561 \nlow-abundance sequence read count table and FASTA files based on the defined cutoff 562 \nvalue. Sequences with low abundances and below threshold were separated and discarded, 563 \nwhile high-abundance sequences were selected for downstream analyses. Following ASV 564 \nextraction for all samples, downstream processing, including sample wise sequence cleaning 565 \nand sorting of FASTA files by specific ASV names, was performed in R using the phytools 566 \n[71], microseq [72], biostrings [73], dplyr [74], ggh4x [75] and tidyverse [76] packages. A 567 \nFASTA file and group file containing corresponding abundance data generated by Mothur 568 \nwere used for downstream analysis. The dataset was standardised to prevent mismatches. 569 \nThe data was used to generate population-specific sequence fasta files, where sequences 570 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n25 \n \nwere replicated according to their abundance values. Biostrings-based function was used to 571 \neach FASTA file to improve data quality. This function removed ambiguous nucleotides (e.g., 572 \n'N') and non-ATCG characters and filtered out sequences shorter than 100 base pairs. The 573 \ncleaned sequences were saved into new FASTA files, and ASVs were named in each sample 574 \nbased on the descending order of sequence read abundance. ASV sequences were 575 \nvisualised using Geneious version 8.0.0 (https://www.geneious.com). Finally, sample files 576 \nare grouped into populations using an R script, and unique sequences were extracted for 577 \neach population for downstream analysis. All R scripts and the reference sequences library 578 \nused for this process are available at the Mendeley data repository 579 \n(https://data.mendeley.com/datasets/822rxwph9t/1).  580 \n Phylogenetic, network and Split Tree analysis 581 \n Phylogenetic trees were generated from unique reference sequences of mt -ND1 and mt-582 \nCOX1 for F. hepatica and F. gigantica, downloaded from NCBI GenBank. The sequences were 583 \naligned using MUSCLE in Geneious v8.0.5. Further, phylogenetic trees were constructed using 584 \nthe Neighbour-Joining method [77]. The evolutionary distances were computed using the 585 \nMaximum Composite Likelihood method [78] in MEGA11 [79] with a bootstrap value of 2000 586 \n[80]. 587 \nSplit trees were generated using SplitTrees4 CE 6.0.0 [81], employing the HKY85 Distance 588 \n[82] and Neighbor Net method [83,84]. The most appropriate nucleotide substitution model 589 \nfor HKY85 Distance was identified using jModelTest 12.2.0 [85]. Split topology tree was 590 \ngenerated using UPGMA method [86], and with the 1000 Bootstraps [80]. Moreover, the 591 \nMedian Joining Network tree, nucleotide diversity, Tajima D, and AMOVA analysis were 592 \nperformed using popart-1.7 [87,88]. AMOVA analysis was further confirmed using Arlequin 593 \n[89]. 594 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n26 \n \nData analysis 595 \nPie charts, frequency distributions, and analyses of predominant and rare ASV patterns, as 596 \nwell as the proportional distribution of ASVs within each county and across the population, 597 \nwere generated to characterise F. hepatica populations in sheep and cattle in R using 598 \npackages readxl [90], ggplot2 [91], dplyr [74], and tidyr [92]. Location data points from 599 \nconfirmed positive collection sites were plotted on the UK map. Geographic coordinates 600 \n(longitude and latitude) for each site were taken from Google Maps (Supplementary Table 601 \n7). Mapping was performed using spatial data sourced from the UK Data Service, including 602 \nCensus Support Digitised Boundary Data (1840–present) and Postcode Directories (1980–603 \npresent), which allowed for the accurate visualisation of ASV distribution patterns across the 604 \nUK. All data analysis and visualisations were performed in R version 4.3.3 (https://cran.r-605 \nproject.org/). 606 \nReferences  607 \n1. Mas-Coma, S., Valero, M. A. & Bargues, M. D. Human and animal fascioliasis: origins and 608 \nworldwide evolving scenario. Clin. Microbiol. Rev. 35, e0008819 (2022). 609 \n2. Mas-Coma, S., Valero, M. A. & Bargues, M. D. Fascioliasis. Adv. Exp. Med. Biol. 1154, 71–103 610 \n(2019). 611 \n3. Novobilský, A. et al. Transmission patterns of Fasciola hepatica to ruminants in Sweden. Vet. 612 \nParasitol. 203, 276–286 (2014). 613 \n4. Dreyfuss, G. & Rondelaud, D. Fasciola hepatica: a study of the shedding of cercariae from 614 \nLymnaea truncatula raised under constant conditions of temperature and photoperiod. Parasite 615 \nParis Fr. 1, 401–404 (1994). 616 \n5. Pfukenyi, D. M. & Mukaratirwa, S. A retrospective study of the prevalence and seasonal 617 \nvariation of Fasciola gigantica in cattle slaughtered in the major abattoirs of Zimbabwe between 618 \n1990 and 1999. Onderstepoort J. Vet. Res. 71, 181–187 (2004). 619 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n27 \n \n6. Hotez, P. J. et al. The Global Burden of Disease Study 2010: Interpretation and Implications for 620 \nthe Neglected Tropical Diseases. PLoS Negl. Trop. Dis. 8, e2865 (2014). 621 \n7. Sweeney, J. et al. Climate change refining the impacts for Ireland: Strive Report (2001-CD-C3-622 \nM1) ISBN: 978-1-84095-297-1. (2008). 623 \n8. Caminade, C., van Dijk, J., Baylis, M. & Williams, D. Modelling recent and future climatic 624 \nsuitability for fasciolosis in Europe. Geospatial Health 9, 301–308 (2015). 625 \n9. Rapsch, C. et al. An interactive map to assess the potential spread of Lymnaea truncatula and 626 \nthe free-living stages of Fasciola hepatica in Switzerland. Vet. Parasitol. 154, 242–249 (2008). 627 \n10. Howell, A. K. & Williams, D. J. L. The epidemiology and control of liver flukes in cattle and sheep. 628 \nVet. Clin. North Am. Food Anim. Pract. 36, 109–123 (2020). 629 \n11. Kenyon, F., Sargison, N. D., Skuce, P. J. & Jackson, F. Sheep helminth parasitic disease in south 630 \neastern Scotland arising as a possible consequence of climate change. Vet. Parasitol. 163, 293–631 \n297 (2009). 632 \n12. Rehman, Z. U. et al. Genetic diversity and multiplicity of infection in Fasciola gigantica isolates of 633 \nPakistani livestock. Parasitol. Int. 76, 102071 (2020). 634 \n13. Beesley, N. J., Williams, D. J. L., Paterson, S. & Hodgkinson, J. Fasciola hepatica demonstrates 635 \nhigh levels of genetic diversity, a lack of population structure and high gene flow: possible 636 \nimplications for drug resistance. Int. J. Parasitol. 47, 11–20 (2017). 637 \n14. Hodgkinson, J., Cwiklinski, K., Beesley, N. J., Paterson, S. & Williams, D. J. L. Identification of 638 \nputative markers of triclabendazole resistance by a genome-wide analysis of genetically 639 \nrecombinant Fasciola hepatica. Parasitol. 140, 1523–1533 (2013). 640 \n15. Kelley, J. M. et al. Current Threat of Triclabendazole Resistance in Fasciola hepatica. Trends 641 \nParasitol. 32, 458–469 (2016). 642 \n16. Vilas, R., Vázquez-Prieto, S. & Paniagua, E. Contrasting patterns of population genetic structure 643 \nof Fasciola hepatica from cattle and sheep: implications for the evolution of anthelmintic 644 \nresistance. Infect. Genet. Evol. J. Mol. Epidemiol. Evol. Genet. Infect. Dis. 12, 45–52 (2012). 645 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n28 \n \n17. Sargison, N. D., Shahzad, K., Mazeri, S. & Chaudhry, U. A high throughput deep amplicon 646 \nsequencing method to show the emergence and spread of Calicophoron daubneyi rumen fluke 647 \ninfection in United Kingdom cattle herds. Vet. Parasitol. 268, 9–15 (2019). 648 \n18. Itagaki, T., Kikawa, M., Terasaki, K., Shibahara, T. & Fukuda, K. Molecular characterization of 649 \nparthenogenic Fasciola sp. in Korea on the basis of DNA sequences of ribosomal ITS1 and 650 \nmitochondrial NDI gene. J. Vet. Med. Sci. 67, 1115–1118 (2005). 651 \n19. Itagaki, T. et al. Genetic characterization of parthenogenic Fasciola sp. in Japan on the basis of 652 \nthe sequences of ribosomal and mitochondrial DNA. Parasitol. 131, 679–685 (2005). 653 \n20. Chaudhry, U. et al. Molecular evidence shows that the liver fluke Fasciola gigantica is the 654 \npredominant Fasciola species in ruminants from Pakistan. J. Helminthol. 90, 206–213 (2016). 655 \n21. Rehman, Z. U. et al. High-throughput sequencing of Fasciola spp. shows co-infection and 656 \nintermediate forms in Balochistan, but only Fasciola gigantica in the Punjab province of 657 \nPakistan. Infect. Genet. Evol. 94, 105012 (2021). 658 \n22. Mogha, L. et al. Genetic diversity and population structure of Fasciola gigantica isolated from 659 \ncattle in Malawi. Vet. Res. Commun. 49, 157 (2025). 660 \n23. Cwiklinski, K. et al. Characterisation of a novel panel of polymorphic microsatellite loci for the 661 \nliver fluke, Fasciola hepatica, using a next generation sequencing approach. Infect. Genet. Evol. 662 \n32, 298–304 (2015). 663 \n24. Ellegren, H. Microsatellites: simple sequences with complex evolution. Nat. Rev. Genet. 5, 435–664 \n445 (2004). 665 \n25. Stallings, R. L. et al. Evolution and distribution of (GT)n repetitive sequences in mammalian 666 \ngenomes. Genomics 10, 807–815 (1991). 667 \n26. Johnson, P. C. D. et al. Abundant variation in microsatellites of the parasitic nematode 668 \nTrichostrongylus tenuis and linkage to a tandem repeat. Mol. Biochem. Parasitol. 148, 210–218 669 \n(2006). 670 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n29 \n \n27. Hodel, R. G. J. et al. The report of my death was an exaggeration: A review for researchers using 671 \nmicrosatellites in the 21st century1. Appl. Plant Sci. 4, apps.1600025 (2016). 672 \n28. Whitlock, M. C. G’ST and D do not replace FST. Mol. Ecol. 20, 1083–1091 (2011). 673 \n29. Jakobsson, M., Edge, M. D. & Rosenberg, N. A. The relationship between F(ST) and the frequency 674 \nof the most frequent allele. Genetics 193, 515–528 (2013). 675 \n30. Hoffman, J. I. & Amos, W. Microsatellite genotyping errors: detection approaches, common 676 \nsources and consequences for paternal exclusion. Mol. Ecol. 14, 599–612 (2005). 677 \n31. Ballard, J. W. O. & Whitlock, M. C. The incomplete natural history of mitochondria. Mol. Ecol. 13, 678 \n729–744 (2004). 679 \n32. Schwartz, M. & Vissing, J. Paternal inheritance of mitochondrial DNA. N. Engl. J. Med. 347, 576–680 \n580 (2002). 681 \n33. Bogenhagen, D. & Clayton, D. A. The number of mitochondrial deoxyribonucleic acid genomes in 682 \nmouse L and human HeLa cells: quantitative isolation of mitochondrial deoxyribonucleaic acid. J. 683 \nBiol. Chem. 249, 7991–7995 (1974). 684 \n34. Galtier, N., Nabholz, B., Glémin, S. & Hurst, G. D. D. Mitochondrial DNA as a marker of molecular 685 \ndiversity: a reappraisal. Mol. Ecol. 18, 4541–4550 (2009). 686 \n35. Hagström, E., Freyer, C., Battersby, B. J., Stewart, J. B. & Larsson, N.-G. No recombination of 687 \nmtDNA after heteroplasmy for 50 generations in the mouse maternal germline. Nucleic Acids 688 \nRes. 42, 1111–1116 (2014). 689 \n36. Rehman, Z. U. et al. Genetic diversity and multiplicity of infection in Fasciola gigantica isolates of 690 \nPakistani livestock. Parasitol. Int. 76, 102071 (2020). 691 \n37. Kiær, L. P. et al. Spontaneous gene flow and population structure in wild and cultivated chicory, 692 \nCichorium intybus L. Genet. Resour. Crop Evol. 56, 405–419 (2009). 693 \n38. Moudgil, A. D. & Nehra, A. K. Mitochondrial genetic markers based phylogenetic analyses of 694 \nHyalomma dromedarii Koch, 1844 (Acari: Ixodidae). J. Genet. Eng. Biotechnol. 23, 100460 695 \n(2025). 696 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n30 \n \n39. Amor, N., Farjallah, S., Merella, P., Alagaili, A. N. & Mohammed, O. B. Multilocus approach 697 \nreveals discordant molecular markers and corridors for gene flow between North African 698 \npopulations of Fasciola hepatica. Vet. Parasitol. 278, 109035 (2020). 699 \n40. Shoriki, T. et al. Molecular phylogenetic identification of Fasciola flukes in Nepal. Parasitol. Int. 700 \n63, 758–762 (2014). 701 \n41. Salim, B. et al. Comprehensive mitochondrial genomics of Fasciola gigantica from Sudan: 702 \ninsights into genetic diversity, evolutionary dynamics, and host adaptation. Front. Vet. Sci. 12, 703 \n(2025). 704 \n42. Elliott, T. et al. Evidence for high genetic diversity of NAD1 and COX1 mitochondrial haplotypes 705 \namong triclabendazole resistant and susceptible populations and field isolates of Fasciola 706 \nhepatica (liver fluke) in Australia. Vet. Parasitol. 200, 90–96 (2014). 707 \n43. Walker, S. M. et al. Evidence for multiple mitochondrial lineages of Fasciola hepatica (liver fluke) 708 \nwithin infrapopulations from cattle and sheep. Parasitol. Res. 101, 117–125 (2007). 709 \n44. Hecker, A. S., Raulf, M.-K., König, S., May, K. & Strube, C. Population genetic analysis of the liver 710 \nfluke Fasciola hepatica in German dairy cattle reveals high genetic diversity and associations 711 \nwith fluke size. Parasit. Vectors 18, 51 (2025). 712 \n45. Calvani, N. E. D. & Šlapeta, J. Fasciola species introgression: just a fluke or something more? 713 \nTrends Parasitol. 37, 25–34 (2021). 714 \n46. Cho, Y. et al. Comparative analysis of Sanger and next generation sequencing methods for 16S 715 \nrDNA analysis of post-mortem specimens. Aust. J. Forensic Sci. 51, 426–445 (2019). 716 \n47. Zahid, O., Chaudhry, U. & Sargison, N. D. Development and field application of metabarcoding-717 \nadapted mt-ND4 markers shows substantial gene flow and varying local pressures on 718 \nHaemonchus contortus and Teladorsagia circumcincta populations in the UK. PLOS ONE 20, 719 \ne0327254 (2025). 720 \n48. Xifeng, W. et al. Molecular characteristics and genetic diversity of Fasciola hepatica from sheep 721 \nin Xinjiang, China. J. Vet. Res. 66, 199–207 (2022). 722 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n31 \n \n49. Walker, S. M. et al. Population dynamics of the liver fluke, Fasciola hepatica: the effect of time 723 \nand spatial separation on the genetic diversity of fluke populations in the Netherlands. Parasitol. 724 \n138, 215–223 (2011). 725 \n50. Chaouadi, M. et al. New insights into the genetic variability of Fasciola hepatica (trematoda) in 726 \nAlgeria and relationships with other geographic regions revealed by mitochondrial DNA. 727 \nHelminthologia 59, 152–164 (2022). 728 \n51. Garcia-Corredor, D. et al. Molecular characterization of Fasciola hepatica in endemic regions of 729 \nColombia. Front. Vet. Sci. 10, (2023). 730 \n52. Thang, T. N. et al. Genetic diversity of Fasciola hepatica in Spain and Peru. Parasitol. Int. 76, 731 \n102100 (2020). 732 \n53. Cabrera, G. et al. Molecular characterization of Fasciola hepatica obtained from cattle and horse 733 \nin Central Chile. Vet. Parasitol. Reg. Stud. Rep. 56, 101130 (2024). 734 \n54. Alvi, M. A. et al. Genetic variation and population structure of Fasciola hepatica: an in silico 735 \nanalysis. Parasitol. Res. 122, 2155–2173 (2023). 736 \n55. Modabbernia, G., Meshgi, B. & Kinsley, A. C. Climatic variations and Fasciola: a review of impacts 737 \nacross the parasite life cycle. Parasitol. Res. 123, 1–14 (2024). 738 \n56. Jones, R. A. et al. Identification of factors associated with Fasciola hepatica infection risk areas 739 \non pastures via an environmental DNA survey of Galba truncatula distribution using droplet 740 \ndigital and quantitative real-time PCR assays. Environ. DNA 6, e371 (2024). 741 \n57. Mas-Coma, S., Valero, M. A. & Bargues, M. D. Climate change effects on trematodiases, with 742 \nemphasis on zoonotic fascioliasis and schistosomiasis. Vet. Parasitol. 163, 264–280 (2009). 743 \n58. Haydock, L. A. J., Pomroy, W. E., Stevenson, M. A. & Lawrence, K. E. A growing degree-day model 744 \nfor determination of Fasciola hepatica infection risk in New Zealand with future predictions 745 \nusing climate change models. Vet. Parasitol. 228, 52–59 (2016). 746 \n59. Pritchard, G. C., Forbes, A. B., Williams, D. J. L., Salimi-Bejestani, M. R. & Daniel, R. G. Emergence 747 \nof fasciolosis in cattle in East Anglia. Vet. Rec. 157, 578–582 (2005). 748 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n32 \n \n60. Green, D. m, Kiss, I. z & Kao, R. r. Modelling the initial spread of foot-and-mouth disease through 749 \nanimal movements. Proc. R. Soc. B Biol. Sci. 273, 2729–2735 (2006). 750 \n61. Volkova, V. V., Howey, R., Savill, N. J. & Woolhouse, M. E. J. Sheep movement networks and the 751 \ntransmission of infectious diseases. PLoS ONE 5, e11185 (2010). 752 \n62. Semyenova, S. K. et al. Genetic differentiation in eastern European and western Asian 753 \npopulations of the liver fluke, Fasciola hepatica, as revealed by mitochondrial NAD1 and COX1 754 \ngenes. J. Parasitol. 92, 525–530 (2006). 755 \n63. Department for Environment, Food & Rural Affairs, UK. Graze with livestock to maintain and 756 \nimprove habitats – Farming. https://defrafarming.blog.gov.uk/graze-with-livestock-to-maintain-757 \nand-improve-habitats/ (2025). 758 \n64. Javanmard, E. et al. Multigene typing and phylogenetic analysis of Fasciola from endemic foci in 759 \nIran. Infect. Genet. Evol. 80, 104202 (2020). 760 \n65. Ashrafi, K. The status of human and animal fascioliasis in Iran: A narrative review article. Iran. J. 761 \nParasitol. 10, 306–328 (2015). 762 \n66. Moghaddam, A. S. et al. Human and animal fascioliasis in Mazandaran province, northern Iran. 763 \nParasitol. Res. 94, 61–69 (2004). 764 \n67. Cwiklinski, K. et al. The Fasciola hepatica genome: gene duplication and polymorphism reveals 765 \nadaptation to the host environment and the capacity for rapid evolution. Genome Biol. 16, 71 766 \n(2015). 767 \n68. Abbas, M. et al. Development of a qPCR assay for Fasciola spp. identification and deep amplicon 768 \nsequencing method for differentiation of fluke species in UK livestock. PLOS Negl Trop Dis 20, 769 \ne0014006 (2026). 770 \n69. Yasein, G. et al. A novel metabarcoded deep amplicon sequencing tool for disease surveillance 771 \nand determining the species composition of Trypanosoma in cattle and other farm animals. Acta 772 \nTrop. 230, 106416 (2022). 773 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n33 \n \n70. Schloss, P. et al. Introducing mothur: open-source, platform-independent, community-774 \nsupported software for describing and comparing microbial communities. Appl. Environ. 775 \nMicrobiol. 75, (2009). 776 \n71. Revell, L. J. phytools 2.0: an updated R ecosystem for phylogenetic comparative methods (and 777 \nother things). PeerJ 12, e16505 (2024). 778 \n72. Snipen, L. & Liland, K. H. microseq: basic biological sequence handling. 779 \nhttps://doi.org/https://cran.r-project.org/web/packages/microseq/index.html (2025). 780 \n73. Pagès, H., Aboyoun, P., Gentleman, R. & DebRoy, S. Biostrings: Efficient manipulation of 781 \nbiological strings. R Package Version 2, 10–18129 (2019). 782 \n74. Wickham, H. et al. dplyr: A grammar of data manipulation. https://doi.org/https://cran.r-783 \nproject.org/web/packages/dplyr/index.html (2023). 784 \n75. Brand, T. van den. ggh4x: Hacks for ‘ggplot2’. https://doi.org/https://cran.r-785 \nproject.org/web/packages/ggh4x/index.html (2025). 786 \n76. Wickham, H. et al. Welcome to the tidyverse. J. Open Source Softw. 4, 1686 (2019). 787 \n77. Saitou, N. & Nei, M. The neighbor-joining method: a new method for reconstructing 788 \nphylogenetic trees. Mol. Biol. Evol. 4, 406–425 (1987). 789 \n78. Tamura, K., Nei, M. & Kumar, S. Prospects for inferring very large phylogenies by using the 790 \nneighbor-joining method. Proc. Natl. Acad. Sci. 101, 11030–11035 (2004). 791 \n79. Tamura, K., Stecher, G. & Kumar, S. MEGA11: Molecular evolutionary genetics analysis version 792 \n11. Mol. Biol. Evol. 38, 3022–3027 (2021). 793 \n80. Felsenstein, J. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39, 794 \n783–791 (1985). 795 \n81. Huson, D. H. & Bryant, D. Application of phylogenetic networks in evolutionary studies. Mol. 796 \nBiol. Evol. 23, 254–267 (2006). 797 \n82. Hasegawa, M., Kishino, H. & Yano, T. Dating of the human-ape splitting by a molecular clock of 798 \nmitochondrial DNA. J. Mol. Evol. 22, 160–174 (1985). 799 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n34 \n \n83. Bryant, D. & Moulton, V. Neighbor-net: An agglomerative method for the construction of 800 \nphylogenetic networks. Mol. Biol. Evol. 21, 255–265 (2004). 801 \n84. Bryant, D. & Huson, D. H. NeighborNet: improved algorithms and implementation. Front. 802 \nBioinforma. 3, (2023). 803 \n85. Posada, D. jModelTest: phylogenetic model averaging. Mol. Biol. Evol. 25, 1253–1256 (2008). 804 \n86. Sokal, R. R. & Michener, C. D. A statistical method for evaluating systematic relationships. 805 \n(1958). 806 \n87. Bandelt, H. J., Forster, P. & Röhl, A. Median-joining networks for inferring intraspecific 807 \nphylogenies. Mol. Biol. Evol. 16, 37–48 (1999). 808 \n88. Leigh, J. W. & Bryant, D. popart: full-feature software for haplotype network construction. 809 \nMethods Ecol. Evol. 6, 1110–1116 (2015). 810 \n89. Excoffier, L., Laval, G. & Schneider, S. Arlequin (version 3.0): An integrated software package for 811 \npopulation genetics data analysis. Evol. Bioinforma. Online 1, 47–50 (2007). 812 \n90. Wickham, H. et al. readxl: Read Excel Files. https://doi.org/https://cran.r-813 \nproject.org/web/packages/readxl/index.html (2025). 814 \n91. CRAN: ggplot2 citation info. https://cran.r-project.org/web/packages/ggplot2/citation.html. 815 \n92. Wickham, H. et al. tidyr: Tidy Messy Data. https://doi.org/https://cran.r-816 \nproject.org/web/packages/tidyr/index.html (2025). 817 \nAcknowledgements 818 \nPart of this work was carried out using computational HPC facilities and support provided by 819 \nthe Research Computing Services team within IT Services at the University of Surrey, 820 \nspecifically the Eureka2 HPC cluster 821 \n(https://docs.pages.surrey.ac.uk/research_computing/hpc/clusters/eureka2.html). 822 \nThis research was funded by the UK Research and Innovation (UKRI), Biotechnology and 823 \nBiological Sciences Research Council (BBSRC) through the FoodBioSystems Doctoral Training 824 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n35 \n \nProgramme (BB/T008776/1) and by the Sir Halley Stewart Trust (3153). For Open Access, the 825 \nauthors have applied a Creative Commons Attribution (CC BY) public copyright license to any 826 \nAuthor Accepted Manuscript version arising from this submission.  827 \nWe sincerely acknowledge all farmers and registered veterinary practitioners in the UK and 828 \nespecially Dr. Iñaki Deza -Cruz (The Royal (Dick) School of Veterinary Studies and The Roslin 829 \nInstitute, The University of Edinburgh, Easter Bush Veterinary Centre, Midlothian, EH25 9RG) 830 \nfor reading the manuscript and sample collection. Dr. Sai Fingerhood (Department of 831 \nVeterinary Pathology, University of Nottingham, UK), and Dr. Mark W. Robinson (School of 832 \nBiological Sciences, Queen's University Belfast, UK) for their valuable assistance in sample 833 \ncollection. 834 \nContributions 835 \nMuhammad Abbas: conceptualisation, investigation, methodology, bioinformatics, 836 \nvalidation, visualisation, data curation and analysis, writing original draft, review and editing; 837 \nKezia Kozel: methodology, writing review and editing; Nick Selemetas: writing review and 838 \nediting, supervision; Olukayode Daramola: writing review and editing , supervision; Eric R 839 \nMorgan: conceptualisation, funding acquisition , supervision, writing review and editing; 840 \nUmer Chaudhry: conceptualisation , writing review and editing, supervision; Martha Betson: 841 \nconceptualisation, writing review and editing, supervision, funding acquisition, project 842 \nadministration. 843 \nEthical statement 844 \nNon-invasive collection of faecal samples was approved by the NASPA (Non-Animal 845 \nScientific Procedures Act) sub-committee of AWERB, University of Surrey, UK, under the 846 \nreference NASPA-2122-04 for the project \"Developing Novel Rapid Diagnostics for 847 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n36 \n \nNeglected Parasitic Diseases.\" Adult F. hepatica were collected at licensed slaughterhouses 848 \nand through post-mortem examination. Completion of a University of Surrey SAGE-AR (ID 849 \n638929-638920-101535552) indicated that no formal ethical approval was required for 850 \nadult fluke sampling.  851 \nSupplementary information 852 \nSupplementary Fig. 1 to 4 853 \nSupplementary Files. 1 to 2 854 \nSupplementary Tables 1 to 7 855 \nRights and permissions 856 \nAll sequencing data reported in the paper are available under NCBI BioProject ID 857 \nPRJNA1402908 and accession numbers : SAMN54606237 -SAMN54606325, PX861700-858 \nPX861710 and PX902280-PX902290. 859 \nIn addition, sequence data, R script, and codes are available at  the Mendeley da tabase 860 \nhttps://data.mendeley.com/datasets/822rxwph9t/1 861 \nAll other data are reported in the paper and associated supplementary material. 862 \nFunding 863 \nMuhammad Abbas received funding from the UK Research and Innovation (UKRI), 864 \nBiotechnology and Biological Sciences Research Council (BBSRC) through the FoodBioSystems 865 \nDoctoral Training Programme for project ID FBS2022 titled \"New tools for sustainable control 866 \nof liver fluke in ruminants\" Grant Ref: BB/T008776/1. Further, this research was funded by 867 \nthe Sir Halley Stewart Trust under the project \"Rapid Diagnostics for Neglected Parasites. 868 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n37 \n \nCompeting Interest 869 \nThe authors declare that no financial interests or personal relationships could have influenced 870 \nthe work reported in this paper.871 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n38 \n \n \nFig. 1. The relative frequencies of mt-ND1 and mt-COX1 ASVs in F. hepatica from sheep (S) \nand cattle (C) across 17 counties in the UK. (a) mt-ND1 ASVs and (b) mt-COX1 ASVs in 40 and \n42 F. hepatica  populations, respectively. Each pie chart represents a distinct population, \noriginating either from adult worms (indicated with an asterisk *) or eggs purified from faeces. \nIndividual ASVs are distinguished by different colours within the charts . The size and \ncomposition of each pie chart reflect the proportional distribution of ASVs within the \nrespective population. Furthermore, each population is mapped to its geographical collection \nsite, providing a visual representation of ASV diversity and distribution across the UK. \n \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n39 \n \n \nFig. 2. Network tree and clustering of all ASVs based on region and host (a) mt-ND1 \nNeighbour Net tree in Split tree. Each pie chart shows regions represented by different \ncolours, with representative ASVs displayed. The pie chart represents the ASV distribution, \nand its frequency in all populations found in the region. The branch lengths were calculated \nusing the HKY85 Distance method, as determined to be best by jModeltest 2.1.10. (b) Each \npie chart presents the countries of the UK, represented by different colours: England cattle \n(red), England sheep (light red), Northern Ireland sheep (yellow), and Scotland cattle (blue), \nand Scotland sheep (light blue), where representative ASVs were recorded. The pie chart \nshows the ASV distribution and its read frequency across all populations in the countries. \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n40 \n \n \nFig. 3. Principal Component Analysis (PCA) plot and topology tree of all ASVs based on \nregion and host. (a) PCA of F. hepatica populations based on 11 mt-ND1 ASV sequence \nabundance. Each point represents a population, with symbols representing the different \nregions and colour the host species. The axes represent the first two principal components \n(PC1 and PC2), which explain 18.31% and 15.18% of the variance, respectively. (b) Split \ntopology tree of mt-ND1 with the UPGMA method. The pie chart circles in the tree \nrepresent the frequency of ASVs sequence reads in all populations.     \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n41 \n \n \nFig. 4. Network tree and clustering of all ASVs based on region and host (a) mt-COX1 \nNeighbor Net tree in Split tree. Each pie chart shows regions represented by different \ncolours, with representative ASVs recorded. The pie chart circle represents the ASV \ndistribution, and its sequence reads frequency in all populations found in the region. The \nbranch lengths were calculated using the HKY85 Distance method, as determined to be best \nby jModeltest 2.1.10. (b) Each pie chart shows the countries of the UK, represented by \ndifferent colours: England cattle (red), England sheep (light red), Northern Ireland sheep \n(yellow), and Scotland cattle (blue) and Scotland sheep (light blue), with representative ASVs \nrecorded. The pie chart shows the ASV distribution and read frequency across all \npopulations in the countries.  \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n42 \n \n \nFig. 5. Principal Component Analysis (PCA) plot and topology tree of all mt-COX1 ASVs (a) \nPCA of F. hepatica populations based on 11 mt-COX1 ASV sequence abundance. Each point \nrepresents a population, with symbols representing the different regions and colour the \nhost species. The axes represent the first two principal components (PC1 and PC2), which \nexplain 19.07% and 15.69% of the variance, respectively. (b) Split topology tree of mt-COX1 \nusing the UPGMA method. The pie charts represent the frequency of ASVs' sequence reads \nin all populations.    \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n43 \n \nTable 1. AMOVA for 11 ASVs of mt-ND1   \nSource of variation df Sum of squares Variance (σ²) % variation \nAmong groups1 16 10.386 0.017 1.3136% \nAmong populations within groups2 23 14.502 –0.508 –39.17% \nWithin populations 54 96.633 1.790 137.853% \nTotal 93 121.521 1.298 100% \nNote: The populations are grouped as follows: 1 all populations observed in a county were \ncategorised as a group; 2 populations found within the group. The negative variance \ncomponent observed in the AMOVA indicates the absence of genetic structure and should \nbe considered as zero. \n \n \nTable 2. AMOVA for 11 AVS of mt-COX1   \nSource of Variation df Sum of Squares Variance (σ²) % Variation \nAmong groups1 16 29.009 0.040 0.74717 % \nAmong populations within groups2 25 50.794 -2.153 -40.16418 % \nWithin populations 68 508.133 7.473 139.41701 % \nTotal 109 587.936 5.360 100 % \nNote: The populations are grouped as follows: 1 all populations observed in a county were \ncategorised as a group; 2 populations found within the group. The negative variance \ncomponent observed in the AMOVA indicates the absence of genetic structure and should \nbe considered as zero. \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint \n\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted April 1, 2026. ; https://doi.org/10.64898/2026.04.01.715781doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}