Comparative Mammalian Genomics

National Human Genome Research Institute

Investigators

Linked publications, trials & patents

Abstract

Canine Genetics The tremendous phenotypic diversity of modern dog breeds represents the end point of a 20,000-year experiment. Each breed has undergone strong artificial selection for morphologic and behavioral traits to create populations with unique traits. As a result, there is strong phenotypic homogeneity within breeds, often leading to breed-enriched disease predispositions. These observations are explored in depth by the Ostrander lab. Canine Whole Genome Sequence We have continued our leadership role with Dog10K, an International Consortium involving 18 labs which aims to generate whole genome sequence (WGS) across the Canidae. Thus far, about 3000 canines are completed. Sequence from samples contributed by the Ostrander lab and most others have been made public. The goal is to sequence (20x) multiple dogs from each of the 300 registered breeds from the U.S., Africa, Europe, and Asia as well as mixed breeds, village dogs, and wild canids (Ostrander et al., 2019). As a result of this project, we have recently published a chromosome-level highly contiguous genome assembly of the initial reference Boxer, created with long-read technologies that increase sequence contiguity by >100-fold, close >23,000 gaps from the CanFam3.1 reference assembly and improve gene annotation by identifying >1200 new protein-coding transcripts (accession GCF_000002285.5) (Jagannathan et al., 2021). Phylogenetic Studies Using genetic distance measures and genome-wide haplotype sharing, we previously observe evidence of phenotypic, behavioral, and geographic patterns of breed development (Parker et al., 2017). We now expand our studies to develop a neighbor-joining cladogram of 357 breeds with dogs from 31 countries. This work enables us to accurately trace the relationships between breeds, setting the stage for studies of newly created breeds, identifying source populations for common breed-types and perform disease allele tracking. We also explored the process of breed formation by studying development of the Australian labradoodle (ALBD) (Ali et al., 2020), one of many new designer breeds. Our haplotype and admixture analyses show that at the genetic level the ALBD is predominantly poodle, but that it retains some of the desired hypo allergenicity-associated traits. The study serves as a blueprint to tracking ongoing breed formation and trait migration. Morphology We previously documented over 91 million canine genome variants, thus creating the largest catalog of genomic variation for a companion animal species at that time (Plassais et al., 2019). Using both selective sweep analyses and genome wide association studies (GWAS) we identify strong impact variants associated with 16 phenotypes, including body weight variation, which highlights genes that, when mutated in humans, contribute to fatty acid metabolism, obesity and metabolic syndrome. Other variants, such as those controlling ear size and shape, produce hearing disorders when mutated in humans. Other traits are similarly defined (Parker et al., 2019; Whitaker et al., 2020). We thus show that GWAS scans performed with whole genome sequencing are powerful methods for finding genes and variants underlying complex traits, thus expanding the utility of companion animal systems for the study of mammalian biology. Independent of the above we worked with the Marques-Bonet lab in Spain to show that copy number variation (CNV) underlies complex phenotypes in dog breeds (Serres-Armero et al., 2021). We generated whole genome CNV maps for 100 dog breeds which we used to perform the first CNV-based GWAS in canids. We identify 96 loci that display CNV differences across breeds that are statistically associated with previously identified breed-specific morphometrics and disease susceptibilities. Integration of these data with chromatin interaction, long noncoding RNA expression, and single nucleotide variation data highlight a subset of specific loci and genes with potential functional relevance that may explain a portion of trait variation between breeds. Canine Cancer We are mapping risk variants for histiocytic sarcoma (HS) and invasive urothelial cancer (iUC). Regarding the first, more than 16,000 people in the U.S. and 160,000 worldwide die annually from iUC. Building upon our previously published findings showing that a commonly mutated form of the BRAF(V595E) gene is present in more than 85% of canine iUC (Decker et al., 2015), we compared tumors with and without the BRAF(V595E) mutation. We observed differential expression of 398 genes that were largely related to cell cycle, immune response, and membrane transport (Parker et a., 2020). We also identified a new locus which harbors 21 upregulated genes in tumors that lack the BRAF(V595E) mutation. Among the most provocative is NECTIN4, which is strongly upregulated in 60% of human bladder tumors. An anti-Nectin-4 drug conjugate has been recently FDA approved for treatment of iUC. By comparison, HS in humans is a rare, aggressive cancer of dendritic cells and macrophages that accounts for <1% of hematopoietic malignancies in humans. Canine HS is a histologically and clinically similar disease which kills 20%-25% of Bernese Mountain Dogs and Flat Coated Retrievers (FCR). We used a multi-omics approach to identify germline regulatory variants associated in HS and other hematopoietic malignancies in FCR (Evans et al., 2021), identifying PIK3R6 and TNFAIP6 as risk loci. Together these loci account for a striking 35% of disease risk. Variants at the PIK2R6 locus colocalize with susceptibility loci for two other hematopoietic malignancies, hemangiosarcoma and B-cell lymphoma, in the closely related Golden Retriever breed, thus revealing the risk contribution this single locus makes to multiple hematological cancers. Aging Working with the Ideker lab at UCSD (Wang et al., 2020) we demonstrate a quantitative translation of dog-to-human aging by conserved remodeling of the DNA methylome. We mapped and compared common methylation changes experienced by mammalian genomes as they age in mice, humans and dogs. Comparison with humans reveals a nonlinear relationship that translates dog-to-human years with the timing of major physiological milestones. Conserved changes occur primarily in developmental gene networks. We are now working with the Horvath lab at UCLA to study the relationship between aging and body size across breeds. Domestication and Endangered Species A natural outgrowth of this work has been our continued collaborations to better understand the process of domestication. Such studies highlight loci that are particularly malleable, providing insights into the genes that are likely most important in understanding human variation. We are co-leading a study with the Larson group at Oxford to examine genome sequence from hundreds of ancient dogs from the Americas, with the aim of understanding migration of both dogs and people during early settlement. Sequencing of ancient samples has begun at NIH. We recently completed a study of new Guinea Singing Dogs (NGSD), which are identifiable by their unusual vocalizations. Once plentiful on the island of New Guinea, they were presumed to exist only in captivity. We analyzed the first nuclear genomes from a scarce population of Highland Wild Dogs (HWD) discovered during a recent expedition to the highlands of New Guinea. Our genome analysis shows strong similarity between HWD and the genetically homogeneous captive NGSD. Admixture analyses and estimation of shared haplotypes with phylogenetically diverse populations indicates the HWD is a novel population which likely represents the forerunner of the now conservation captive population (Surbakti et al., 2020). Taken together, these data indicate the HWD possesses a distinct potential to aid in the conservation of NGSD.