Distinctive genetic structure and selection patterns in Plasmodium vivax from South Asia and East Africa

Abstract

Despite the high burden of Plasmodium vivax malaria in South Asian countries, the genetic diversity of circulating parasite populations is not well described. Determinants of antimalarial drug susceptibility for P. vivax in the region have not been characterised. Our genomic analysis of global P. vivax (n = 558) establishes South Asian isolates (n = 92) as a distinct subpopulation, which shares ancestry with some East African and South East Asian parasites. Signals of positive selection are linked to drug resistance-associated loci including pvkelch10, pvmrp1, pvdhfr and pvdhps, and two loci linked to P. vivax invasion of reticulocytes, pvrbp1a and pvrbp1b. Significant identity-by-descent was found in extended chromosome regions common to P. vivax from India and Ethiopia, including the pvdbp gene associated with Duffy blood group binding. Our investigation provides new understanding of global P. vivax population structure and genomic diversity, and genetic evidence of recent directional selection in this important human pathogen.

Using deep learning to identify recent positive selection in malaria parasite sequence data

“…Using a leave-one-group-out validation approach, the overall accuracy was 79.7% (standard deviation 17.6%), and the performance was lower when pvmdr1 was omitted (57.1%) and was higher when pvdhfr was left out (100%). This difference is consistent with pvmdr1 residues being strongly associated with chloroquine resistance [ 5 ] and, although, pvdhfr may contribute to SP drug resistance, there are very few published studies that associate genotypes of this locus with anti-folate susceptibility phenotypes [ 6 ]. As with P. falciparum, the trained model had strong performance both in terms of classification accuracy and reduction of binomial loss (Additional file 1 : Figure S6).…”

“…extreme genome GC content), which can impact on the density and accuracy of genomic variant inputs, as well as the final population genomic analysis. Typically, WGS analysis leads to a dense set of well supported variants in robust genomic regions, with the application of calling algorithms incorporating information on known high quality polymorphisms [ 6 ]. Further, highly variable or problematic regions, such as var genes in P. falciparum , are typically removed from analysis [ 46 ].…”

“…For P. falciparum, these included regions around established SNPs in pfcrt (K76T, I356T; chloroquine) , pfdhfr (N51I, C59R, S108N, I164L, S306F) /pfdhps (I431V, S436A, A437G, K540E/N, A581G, S613S) (SP) , pfmdr1 (N86Y; mefloquine, chloroquine) , and pfkelch13 (F446I, Y493H, P574L, R539T, and C580Y; artemisinin) [ 27 ] . For P. vivax , these included regions around some known SNPs in pvdhps (A553G, G383A, S382C/A) / pvdhfr (N50I, F57I/L, S/K58R, T61M, N117T/S) (putative SP) and pvmdr1 (F1076L, Y976F, S698G, S513R; putative chloroquine) [ 4 , 6 ]. This could be considered a relatively small number of training exemplars, which may lead to an increased risk that the implemented machine learning algorithm overfits due to potential artefacts in the training data.…”

“…For P. vivax , the spread of resistance to chloroquine, primaquine, mefloquine, and SP has been reported in various regions of the world [ 4 , 5 ]. The underlying mutations causing resistance for P. vivax are less well defined than for P. falciparum [ 4 – 6 ].…”

DataSheet_1.docx

“…Population genomic analyses of P. vivax populations in Southeast Asia and Oceania found evidence that the pvmrp1 gene is under a strong selection (Pearson et al, 2016;Benavente et al, 2017;Benavente et al, 2021). In South Asian populations, the pvmrp1 gene exhibited high frequencies of two nonsynonymous substitutions (D1393Y and G1419A).…”