Modeling malaria genomics reveals transmission decline and rebound in Senegal

Abstract

To study the effects of malaria-control interventions on parasite population genomics, we examined a set of 1,007 samples of the malaria parasite Plasmodium falciparum collected in Thiès, Senegal between 2006 and 2013. The parasite samples were genotyped using a molecular barcode of 24 SNPs. About 35% of the samples grouped into subsets with identical barcodes, varying in size by year and sometimes persisting across years. The barcodes also formed networks of related groups. Analysis of 164 completely sequenced parasites revealed extensive sharing of genomic regions. In at least two cases we found first-generation recombinant offspring of parents whose genomes are similar or identical to genomes also present in the sample. An epidemiological model that tracks parasite genotypes can reproduce the observed pattern of barcode subsets. Quantification of likelihoods in the model strongly suggests a reduction of transmission from 2006-2010 with a significant rebound in 2012-2013. The reduced transmission and rebound were confirmed directly by incidence data from Thiès. These findings imply that intensive intervention to control malaria results in rapid and dramatic changes in parasite population genomics. The results also suggest that genomics combined with epidemiological modeling may afford prompt, continuous, and cost-effective tracking of progress toward malaria elimination.

Keywords: epidemiology; genomics; malaria.

Fig. 1.

Relatedness among parasite isolates. (A) Network of barcode relatedness based on genetic distances between barcodes (19), in which edge thickness represents degree of identity. The thickest edges connect samples 95.8–100% related (identical or one SNP difference), and the thinnest edges connect samples that are less than 87.5% related (five SNP differences). Colored dots indicate barcodes present three or more times in the samples; gray indicates those present one or two times. (B) Network of sample relatedness based on full sequence data, in which edge thickness represents the fraction of the genome that is identical by descent; node colors correspond to those in A. The red square and circle indicate clusters containing the parents of parasite sample SenT120.11.

Fig. 2.

Size distribution of shared sequence blocks. (A) Size distribution of regions of identity by descent (blocks of shared sequence) between nonidentical pairs of isolates among the sequenced isolates. (B) Parental origin of genomic segments for sample SENT120.11. Colored segments show identity by descent (as assigned by a hidden Markov model) to the two parental types (blue and orange), to both types (black), or to neither (gray).

Fig. 3.

Model output calibrated to observed barcode data. (A) Unique and repeated subsets of barcodes in observed data. (B) Unique and repeated subsets of barcodes in data output from the fitted model.

Fig. 4.

Changing transmission dynamics in the epidemiological model fitted to barcode data. (A) Seasonal changes in relative number of affected individuals and unique barcodes across years. $R_0^a$ is the initial maximum rate of parasite increase, which is assumed to decrease linearly to a minimum rate ($R_0^b$) and then to rebound to a rate $R_0^c$. The curves shown are for the maximum likelihood estimates of the three rates of parasite increase. (B) Log-likelihoods for values of $R_0^b$ versus $R_0^a$. (C) Log-likelihoods for values of $R_0^b$ versus $R_0^c$.

Fig. 5.

Malaria incidence per person, 2006−2013, normalized to that observed in 2006 and fitted to a model with an exponential decrease plus a rebound. (A) Data from Thiès, in which the rebound is statistically significant. (B) Data from all of Senegal excluding Thiès, which shows no significant rebound. Incidence data are from the Senegal National Malaria Control Program (2).