Demographic history and biologically relevant genetic variation of Native Mexicans inferred from whole-genome sequencing

Abstract

Understanding the genetic structure of Native American populations is important to clarify their diversity, demographic history, and to identify genetic factors relevant for biomedical traits. Here, we show a demographic history reconstruction from 12 Native American whole genomes belonging to six distinct ethnic groups representing the three main described genetic clusters of Mexico (Northern, Southern, and Maya). Effective population size estimates of all Native American groups remained below 2,000 individuals for up to 10,000 years ago. The proportion of missense variants predicted as damaging is higher for undescribed (~ 30%) than for previously reported variants (~ 15%). Several variants previously associated with biological traits are highly frequent in the Native American genomes. These findings suggest that the demographic and adaptive processes that occurred in these groups shaped their genetic architecture and could have implications in biological processes of the Native Americans and Mestizos of today.

Fig. 1

Map of Mexico. The figure shows the geographical origin of each NA and unrelated Mestizo individuals included in the present study. The region of Mesoamerica is shaded in gray. TAR (Tarahumara), TEP (Tepehuano), NAH (Nahua), TOT (Totonaca), ZAP (Zapoteca), MAY (Maya), and MES (Mestizo)

Fig. 2

Multidimensional scaling plots and admixture analysis. a MDS plot for components 1 and 2 of 12 Native Americans of Mexico combined with continental (CEU and YRI from HapMap) and NA reference populations. b MDS plot for components 1 and 3, separating the NA populations of Mexico. The 12 sequenced NA samples are shown in gray, and population labels are as described in Supplementary Table 8. c Global ancestry proportions of NA and continental reference populations assuming K = 10. d Global ancestry proportions of the 12 NA individuals assuming K = 10. The NA individuals are displayed North-to-South, and the Mestizo individuals are displayed at the far right. Tar1 and Tar2 (Tarahumara), Tep1 and Tep2 (Tepehuano), Nah1 and Nah2 (Nahua), Tot1 and Tot2 (Totonaca), Zap1 and Zap2 (Zapoteca), May1 and May2 (Maya), Mes1 and Mes2 (Mestizo)

Fig. 3

Inference of population sizes from whole-genome NA sequences. a Effective population size estimates from autosomes of six Native American (NA) individuals and an African, European, and Asian genome from the 1KG project, inferred using pairwise sequentially Markovian coalescent (PSMC) models. b Effective population size estimates from four haplotypes (two phased individuals from each of the six NA and 1KG populations), inferred using multiple sequentially Markovian coalescent (MSMC) models. c Effective population size estimates from eight haplotypes [four phased individuals from each of NA Northern (Tepehuanos and Tarahumaras), NA Southern (Nahuas and Zapotecas) and 1KG populations], inferred using MSMC models

Fig. 4

Enrichment analysis for disease-associated genes. Annotation matrix for disease-associated genes enriched in different populations shown as a heatmap, red indicates disease terms significantly enriched. Disease terms enrichment was analyzed by WebGestalt using the hypergeometric test for enrichment evaluation analysis. P-values were adjusted using Benjamini & Hochberg method. MES (Mestizo), TEP (Tepehuano), ZAP (Zapoteca), MAY (Maya), NAH (Nahua), TOT (Totonaca), TAR (Tarahumara), HPR TAR (high physical resistance)