%0 Journal Article %1 dou2017estimation %A Dou, Jinzhuang %A Sun, Baoluo %A Sim, Xueling %A Hughes, Jason D. %A Reilly, Dermot F. %A Tai, E. Shyong %A Liu, Jianjun %A Wang, Chaolong %D 2017 %I Public Library of Science %J PLOS Genetics %K kinship methods relatedness %N 9 %P 1-29 %R 10.1371/journal.pgen.1007021 %T Estimation of kinship coefficient in structured and admixed populations using sparse sequencing data %U https://doi.org/10.1371/journal.pgen.1007021 %V 13 %X Author summary Inference of genetic relatedness from molecular markers has broad applications in many areas, including quantitative genetics, forensics, evolution and ecology. Classic estimators, however, are not suitable for low-coverage sequencing data, which have high levels of genotype uncertainty and missing data. We evaluate existing methods and describe a new method for kinship estimation using sparse sequencing data. Our method leverages correlations between neighboring markers and models genotype uncertainty in kinship estimators for both homogeneous populations and admixed populations. We show that our method can accurately estimate kinship coefficient even when the sequencing depth is as low as ~0.15X, while existing methods have strong downward bias. Our method can be applied to estimate kinship using sparse off-target data and thus enables control of family structure and estimation of heritability in target sequencing studies, in which the deeply sequenced target regions are often too small to infer genetic relatedness. Even for whole exome sequencing, we show that our method can improve kinship and heritability estimation by including off-target data, compared to conventional analyses solely based on the target regions.

@article{dou2017estimation, abstract = {Author summary Inference of genetic relatedness from molecular markers has broad applications in many areas, including quantitative genetics, forensics, evolution and ecology. Classic estimators, however, are not suitable for low-coverage sequencing data, which have high levels of genotype uncertainty and missing data. We evaluate existing methods and describe a new method for kinship estimation using sparse sequencing data. Our method leverages correlations between neighboring markers and models genotype uncertainty in kinship estimators for both homogeneous populations and admixed populations. We show that our method can accurately estimate kinship coefficient even when the sequencing depth is as low as ~0.15X, while existing methods have strong downward bias. Our method can be applied to estimate kinship using sparse off-target data and thus enables control of family structure and estimation of heritability in target sequencing studies, in which the deeply sequenced target regions are often too small to infer genetic relatedness. Even for whole exome sequencing, we show that our method can improve kinship and heritability estimation by including off-target data, compared to conventional analyses solely based on the target regions.}, added-at = {2020-10-09T15:23:18.000+0200}, author = {Dou, Jinzhuang and Sun, Baoluo and Sim, Xueling and Hughes, Jason D. and Reilly, Dermot F. and Tai, E. Shyong and Liu, Jianjun and Wang, Chaolong}, biburl = {https://www.bibsonomy.org/bibtex/247510d721fdd85a13d2184128beacf10/peter.ralph}, doi = {10.1371/journal.pgen.1007021}, interhash = {d65cabf6fe80f91bd8e35068dc42adb7}, intrahash = {47510d721fdd85a13d2184128beacf10}, journal = {PLOS Genetics}, keywords = {kinship methods relatedness}, month = {09}, number = 9, pages = {1-29}, publisher = {Public Library of Science}, timestamp = {2020-10-09T15:23:18.000+0200}, title = {Estimation of kinship coefficient in structured and admixed populations using sparse sequencing data}, url = {https://doi.org/10.1371/journal.pgen.1007021}, volume = 13, year = 2017 }