Recent genetic connectivity and clinal variation in chimpanzees

Abstract

Much like humans, chimpanzees occupy diverse habitats and exhibit extensive behavioural variability. However, chimpanzees are recognized as a discontinuous species, with four subspecies separated by historical geographic barriers. Nevertheless, their range-wide degree of genetic connectivity remains poorly resolved, mainly due to sampling limitations. By analyzing a geographically comprehensive sample set amplified at microsatellite markers that inform recent population history, we found that isolation by distance explains most of the range-wide genetic structure of chimpanzees. Furthermore, we did not identify spatial discontinuities corresponding with the recognized subspecies, suggesting that some of the subspecies-delineating geographic barriers were recently permeable to gene flow. Substantial range-wide genetic connectivity is consistent with the hypothesis that behavioural flexibility is a salient driver of chimpanzee responses to changing environmental conditions. Finally, our observation of strong local differentiation associated with recent anthropogenic pressures portends future loss of critical genetic diversity if habitat fragmentation and population isolation continue unabated.

Fig. 1. Distribution map of P. troglodytes and PanAf sampling.

The current approximate chimpanzee subspecies ranges, sample collection locations and proposed subspecies geographic barriers. Total number of genotyped individuals and number of sampling locations are listed for each subspecies population. Much of the historical population located between P. t. ellioti and P. t. verus populations have been extirpated, creating an extensive sampling gap in the data. Note, samples collected during nationwide studies in Liberia and Equatorial Guinea were included in our spatially explicit analyses and are indicated here as the geographic centre points of the sampling distribution.

Fig. 2. Linear regressions of genetic distance as a function of geographic distance.

a Linear regressions of genetic distance as a function of geographic distance for all sites with at least six genotyped individuals. Blue dots represent pairwise comparisons involving outlier sites (Mt. Sangbé, Gashaka and Issa, see text). The purple line is fitted to the entire dataset, while the red line is fitted to the dataset excluding the three outlier sampling locations. When excluding the three outlier sites, geographic distance explains 58% of the genetic distance. b Linear regressions of genetic distance as a function of geographic distance for between- and within-subspecies comparisons. Blue dots represent between-subspecies pairwise comparisons with linear regression (blue line). Green and pink dots represent within-subspecies pairwise comparisons with linear regression (brown line) characterized by a noticeably lower y-intercept. Pan troglodytes verus–P. t. verus comparisons (pink dots), which have the lowest genetic diversity of all subspecies, make up 60% of all the within-subspecies comparisons and are largely driving this observed pattern, explaining the source of the stratification in a. c Linear regressions of genetic distance as a function of geographic distance for each subspecies comparison pair. Solid lines represent fitted regressions, and dashed lines enclose 95% confidence intervals. d Estimates (circles) and standard errors (crosses) of intercept and slope for each of the regressions in c.

Fig. 3. Estimated effective migration surfaces (EEMS) at different population scales.

a Map of EEMS for the entire species. Estimated effective migration rates (m) are mean centred on a log₁₀ scale. A value of 1 equates to a tenfold rate increase over population average, ranging from areas of low (brown) to high (blue) m. The intensity of the colours represents the relative difference from the population mean rates. Point diameter is proportional to the number of individuals sampled in a given deme (range = 1–72). Solid black lines indicate areas where the posterior probability of m differing from the mean rate is >95 percent and the dashed lines highlight areas that are >90 percent. These can be interpreted as significant effective ‘barriers’ to migration in brown areas and significant effective ‘corridors’ for migration in the blue areas. Two significant effective barriers were present: one corresponding to Gashaka and Mbe (Nigeria) and another originating from Gishwati (Rwanda) and shared with its nearest neighbours. These are localized areas of high differentiation, and when we excluded Mbe and Gishwati from the dataset, these barriers were no longer significant (Extended Data Fig. 6a, c). Notably, historical effective barriers separating the subspecies’ ranges were not detected. Effective migration rates within much of Pan troglodytes verus were significantly higher than average. b Historical EEMS map of P. t. verus. There was a significant barrier associated with Mt. Sangbé in Côte d’Ivoire. When we removed Mt. Sangbé the barrier was no longer significant, but it was still present, suggesting the possibility of reduced historical gene flow across Côte d’Ivoire. c Historical EEMS map of P. t. ellioti, P. t. troglodytes and P. t. schweinfurthii (collectively ETS). As in a, removing Gishwati resulted in the barrier no longer being significant (Extended data Fig. 6c). Rates of m between the panels are relative and not directly comparable.

Fig. 4. EEMS diversity rates in chimpanzees at different population scales.

a Map of diversity rates (q) for the entire species. Diversity rates are mean centred (population average) on a log₁₀ scale, whereby a value of 1 equates to a tenfold rate increase over population mean rate. Point diameter is proportional to the number of individuals sampled in a given deme (range = 1–72). Solid black lines indicate areas where the posterior probability of q differing from the mean rate is >95 percent and the dash lines highlight areas that are >90 percent. Lighter areas indicate populations where q is lower than the mean rate, and darker areas indicate populations where q is higher. Diversity rates were significantly lower in Pan troglodytes verus, while many of the populations in the P. t. troglodytes and P. t. schweinfurthii ranges had significantly higher rates. b Map of q in P. t. verus. Grebo and Sapo (Liberia) had significantly higher diversity rates than the average rate in P. t. verus and Djouroutou (Côte d’Ivoire) had significantly lower rates. Diversity at Mt. Sangbé (Côte d’Ivoire, white area) is also notably low, but not significant. c Map of q in P. t. ellioti, P. t. troglodytes and P. t. schweinfurthii (collectively ETS). Several sites differed significantly, with P. t. ellioti sites in particular displaying lower diversity rates, but overall q were homogeneous. Rates of q between panels are relative and not directly comparable.