Whole genome sequencing predicts novel human disease models in rhesus macaques

Abstract

Rhesus macaques are an important pre-clinical model of human disease. To advance our understanding of genomic variation that may influence disease, we surveyed genome-wide variation in 21 rhesus macaques. We employed best-practice variant calling, validated with Mendelian inheritance. Next, we used alignment data from our cohort to detect genomic regions likely to produce inaccurate genotypes, potentially due to either gene duplication or structural variation between individuals. We generated a final dataset of >16 million high confidence variants, including 13 million in Chinese-origin rhesus macaques, an increasingly important disease model. We detected an average of 131 mutations predicted to severely alter protein coding per animal, and identified 45 such variants that coincide with known pathogenic human variants. These data suggest that expanded screening of existing breeding colonies will identify novel models of human disease, and that increased genomic characterization can help inform research studies in macaques.

The burden of deleterious variants in a non-human primate biomedical model

“…The vervet resource described here is complementary to those generated in other NHP models. Recent sequencing surveys of research colony rhesus macaques have similarly revealed many PTVs that in some cases match human PTVs with relevant clinical data (Xue et al 2016;Bimber et al 2017). While in that study, no macaque phenotypes were exploited to correlate to observed phenotypes in human, there is potential to do so in the future.…”

“…To our knowledge no studies have attempted to associate human PTVs, broadly, to observed NHP phenotypes, and there is little information, to date, on the extent of overlap between human and NHP PTVs . Among captive rhesus macaques, two studies have shown various degree of shared PTVs with human, depending on the types of variant filters deployed (Xue et al 2016;Bimber et al 2017). In six great ape populations, a range of PTVs was reported, with a strict focus on the stop gain classification (de Valles-Ibáñez et al 2016).…”

mGAP: the macaque genotype and phenotype resource, a framework for accessing and interpreting macaque variant data, and identifying new models of human disease

“…These animals were selected from a multi-generational outbred cohort of macaques bred at the Oregon National Primate Research Center (average kinship 0.0016). Genomic DNA was isolated, used to generate Illumina libraries as previously described, followed by sequencing on an Illumina XTen sequencer [35]. We obtained an average of 704,193,015 paired Illumina reads per animal, with a range of 549,660,386 -1,843,458,162 (Additional file 1: Table S1).…”

Nonhuman Primates and Translational Research: Progress, Opportunities, and Challenges

“…After an initial sequence is obtained, usually from one or two animals, refinement of the sequence assembly is performed based on sequencing the genome and transcriptomes of many more animals. For example, Xue et al (2016) and Bimber et al (2017) have reported sequencing of several additional rhesus, which can be used to further refine the rhesus genome assembly and allow new inferences about rhesus genome organization and potential for identifying disease genes in captive populations.…”

“…Genomic and exomic sequences of NHPs can now be obtained readily, at moderate, but still significant, cost. The ability to readily sequence individual animals has prompted the champions of this technology to suggest large-scale sequencing of many more animals, particularly rhesus at the NPRCs (Bimber et al 2016(Bimber et al , 2017Cornish et al 2016). This could facilitate, for example, identification of potentially deleterious gene variants present in the heterozygous state that can be developed into homozygous animal models of specific human diseases, particularly for Mendelian (single gene) diseases.…”