Genome-wide SNP genotyping of Plasmodium falciparum isolates across Mali reveals major impacts of Pfsa1 and PfCRT K76T selections on parasite populations.

Malaria is a common disease in Mali associated with high morbidity and mortality. To improve malaria control, it is important to evaluate how current interventions affect the genetic make-up of circulating parasites. We analyzed the genomes of recently collected Plasmodium falciparum isolates across Mali to understand transmission dynamics, parasite relatedness, and selection pressures affecting parasite populations. We sequenced clinical isolates (n=458) collected in 2023 from 13 heath districts across Mali using molecular inversion probe (MIP) panels targeting key drug resistance mutations and common SNPs across the P. falciparum genomes. We also MIP sequenced polymorphisms in the human β-globin gene in study participants to assess sickle haemoglobin mutation. We performed complexity of infection (COI), identity-by-descent (IBD), principal component (PCA), selection signal, and haplotype analyses. The prevalence of polygenomic infections ranged from 13.0% in the north, dominated by the Sahara Desert, to 44.0% in the south, with tropical wet and dry climates. Parasites collected from the same district showed low levels of genetic relatedness with a mean IBD of <0.2. There was a network of parasite connectivity between sites across the country, and IBD between districts was inversely correlated with geographical distance separating them ( P=0.006, R=-0.28 ). The strongest selection signal was detected around the chloroquine-resistance transporter ( pfcrt ) gene. PCA in monogenomic samples (n=242) separated a cluster of parasites (n=56) from the main population showing a specific selection signal around acetyl-coA synthetase ( pfacs8) . The ACS8 Y40F mutation, associated with sickle haemoglobin and also known as Pfsa1 , was detected in all the samples from this cluster. Higher prevalence of PfCRT K76T mutation (55.8%, n=104) and sickle cell trait (8.4%, n=83) were measured in samples carrying Pfsa1 compared to wild type (26.6%, n=342 and 1.3%, n=237; respectively, P<0.005 , Chi-square test). Samples carrying both Pfsa1 and PfCRT K76T exhibited higher IBD across chromosomes, and one major haplotype was identified around pfcrt in 80% of K76T mutants, in which 64% of Pfsa1 was detected. Our nationwide genomic profiling of malaria parasites in Mali reveals clear selective sweeps linked to drug pressure and host genetics highly contributing to shaping the genetic make-up of parasite populations, and confirms the association of Pfsa1 with sickle haemoglobin. The Pfsa1 -PfCRT 76T linkage highlights a critical axis of parasite adaptation that warrants close surveillance.