Maximum-likelihood estimation of recent shared ancestry (ERSA)

• Published in: Genome research Vol. 21; no. 5; pp. 768 - 774
• Main Authors: Huff, Chad D, Witherspoon, David J, Simonson, Tatum S, Xing, Jinchuan, Watkins, W Scott, Zhang, Yuhua, Tuohy, Therese M, Neklason, Deborah W, Burt, Randall W, Guthery, Stephen L, Woodward, Scott R, Jorde, Lynn B
• Format: Journal Article
• Language: English
• Published: United States Cold Spring Harbor Laboratory Press 01-05-2011
• Subjects: Chromosome Mapping; DNA - genetics; Family; Genetic Linkage; Humans; Inheritance Patterns - genetics; Likelihood Functions; Models, Genetic; Models, Statistical; Oligonucleotide Array Sequence Analysis; Pedigree; Polymorphism, Single Nucleotide; Principal Component Analysis; Resource; Software
• ISSN: 1088-9051; 1549-5469; 1549-5469
• DOI: 10.1101/gr.115972.110

Accurate estimation of recent shared ancestry is important for genetics, evolution, medicine, conservation biology, and forensics. Established methods estimate kinship accurately for first-degree through third-degree relatives. We demonstrate that chromosomal segments shared by two individuals due to identity by descent (IBD) provide much additional information about shared ancestry. We developed a maximum-likelihood method for the estimation of recent shared ancestry (ERSA) from the number and lengths of IBD segments derived from high-density SNP or whole-genome sequence data. We used ERSA to estimate relationships from SNP genotypes in 169 individuals from three large, well-defined human pedigrees. ERSA is accurate to within one degree of relationship for 97% of first-degree through fifth-degree relatives and 80% of sixth-degree and seventh-degree relatives. We demonstrate that ERSA's statistical power approaches the maximum theoretical limit imposed by the fact that distant relatives frequently share no DNA through a common ancestor. ERSA greatly expands the range of relationships that can be estimated from genetic data and is implemented in a freely available software package.