Population genomics of Plasmodium ovale species in sub-Saharan Africa

Abstract

Plasmodium ovale curtisi (Poc) and Plasmodium ovale wallikeri (Pow) are relapsing malaria parasites endemic to Africa and Asia that were previously thought to represent a single species. Amid increasing detection of ovale malaria in sub-Saharan Africa, we present a population genomic study of both species across the continent. We conducted whole-genome sequencing of 25 isolates from Central and East Africa and analyzed them alongside 20 previously published African genomes. Isolates are predominantly monoclonal (43/45), with their genetic similarity aligning with geography. Pow shows lower average nucleotide diversity (1.8×10^-4) across the genome compared to Poc (3.0×10^-4) (p < 0.0001). Signatures of selective sweeps involving the dihydrofolate reductase gene have been found in both species, as are signs of balancing selection at the merozoite surface protein 1 gene. Differences in the nucleotide diversity of Poc and Pow may reflect unique demographic history, even as similar selective forces facilitate their resilience to malaria control interventions.

Fig. 1. Coverage and mapping among 45 isolates.

The proportion of the corresponding P. ovale reference genome covered by ≥10 reads by DNA enrichment technique (A) and species (B). The proportion of reads mapped to that reference genome by DNA enrichment technique (C) and species (D). For Poc, chromosome 10 was excluded due to incomplete coverage by hybrid capture baits. Five samples from Higgins et al. were not incorporated into this study due to having <30% 10x coverage of the corresponding genome. HC hybrid capture, LDB leukodepleted blood, sWGA selective whole-genome amplification. Source data are provided as a Source Data file.

Fig. 2. Complexity of infection by Plasmodium species.

Median estimated complexity of infection (COI) among 21 Poc isolates, 24 Pow isolates, and 2077 P. falciparum isolates geographically matched to the P. ovale samples by country of origin. The distributions of COI differed significantly among the three species (p < 0.0001) by a Kruskal–Wallis test, with Pow and Poc showing significantly lower COI than P. falciparum (p = 0.0004 and 0.012, respectively) in Dunn’s multiple comparisons tests. The average read depth of coverage for Poc, Pow, and Pf isolates were 44.1, 95.7, and 147.5, respectively. Source data are provided as a Source Data file.

Fig. 3. Nucleotide diversity (π) of orthologous genes and SNP density by functional genomic region among Poc, Pow, and P. falciparum (Pf) isolates.

A Nucleotide diversity (π) per gene among 2008 sets of orthologous genes in monoclonal Poc, Pow, and Pf samples. Boxes denote the 25th, median, and 75th percentiles; whiskers are drawn at the 1st and 99th percentiles. π of 0 was coded as 1 × 10⁻⁵ to plot on a logarithmic scale. Nucleotide diversity was significantly different between orthologues of all three species by two-sided Tukey’s multiple comparisons tests, with Poc orthologues showing higher diversity than orthologues of Pow and P. falciparum, and Pow orthologues also showing lower diversity than those in P. falciparum (p values <0.001, =0.002, and <0.0001, respectively). B Nucleotide diversity (π) per gene among 2911 sets of orthologous genes in geographically matched monoclonal Poc and Pow samples. Boxes denote the 25th, median, and 75th percentiles; whiskers are drawn at the 1st and 99th percentiles. π of 0 was coded as 1 × 10⁻⁵ to plot on a logarithmic scale. Nucleotide diversity was significantly lower among Pow orthologues compared to Poc using a two-sided Wilcoxon’s matched-pair signed rank test (p < 0.0001). C SNP density in different functional regions of the genome among all Pow, Poc, and P. falciparum isolates. SNP single nucleotide polymorphism, kb kilobase. Source data are provided as a Source Data file.

Fig. 4. Principal component analysis of monoclonal P. ovale spp. isolates.

Principal component analysis showing the first two principal components among 20 monoclonal Poc isolates (A) and 23 monoclonal Pow isolates (B) using 4116 and 3189 biallelic SNPs, respectively. Samples are colored by region of country of origin; in the map, parasites from travelers are assigned to the capital city (C). In PC2 of Pow, SNPs within both the ts-dhfr and mrp1 gene were among the top 0.5% of contributors. Source data are provided as a Source Data file.

Fig. 5. Absolute nS_L across the Poc and Pow genome.

Absolute nS_L of 19,205 variants among monoclonal isolates across the Poc genome (A) and 15,744 variants among monoclonal isolates across the Pow genome (B). Individual loci are depicted using alternating shapes between chromosomes for legibility. The dotted line and red color denote the top 0.5% of loci. Putative genetic markers of note within 10,000 bases of these loci are labeled, including ubiquitin transferase (ut), cysteine-rich secretory protein (crisp), cysteine repeat modular protein 2 (crmp2), early transcribed membrane protein (etramp), GPI-anchored micronemal antigen (gama), dihydrofolate reductase-thymidylate synthase (dhfr-ts), apical membrane antigen (ama1), merozoite surface protein 7-like protein (msp7), male development protein 1 (md1), AP2 domain transcription factor G (ap2-g), multidrug resistance-associated protein 2 (mrp2), dynein heavy chain (dhc), AP2 domain transcription factors (ap2), merozoite surface protein 5 (msp5), cysteine repeat modular protein 1 (crmp1), dynein light chain (dlc), 6-cysteine protein (6-cys), sporozoite protein essential for cell traversal 1 (spect1), Plasmodium interspersed repeat protein (pir), and early transcribed membrane protein (etramp). Source data are provided as a Source Data file.

Fig. 6. Extended haplotype homozygosity and haplotype bifurcation at selected variants.

Extended haplotype homozygosity (EHH) and haplotype bifurcation among monoclonal isolates at selected variants near the Pow dhfr-ts gene (A, B), the Poc dhfr-ts gene (C, D), and the Poc mrp2 gene (E, F). EHH and haplotype bifurcation show selective sweeps spanning ~30, ~40, and ~30 kb, respectively, with lineage breakdown occurring first among the unselected allele haplotypes (blue) and then in the selected allele haplotypes (red) as the distance from the focal variant increases.

Fig. 7. Tajima's D across the Poc and Pow genomes.

Tajima’s D in 399,355 and 297,286 300 bp windows in genes across the Poc (A) and Pow (B) genomes among monoclonal isolates. Individual windows are depicted using alternating shapes between chromosomes for legibility. The dotted line and red color denote the top 0.5% of loci. For both species, the top locus with a positive Tajima’s D was located inside the merozoite surface protein 1 (msp1) gene. Source data are provided as a Source Data file.