AD-LIBS: inferring ancestry across hybrid genomes using low-coverage sequence data.
BackgroundInferring the ancestry of each region of admixed individuals' genomes is useful in studies ranging from disease gene mapping to speciation genetics. Current methods require high-coverage genotype data and phased reference panels, and are therefore inappropriate for many data sets. We...
Main Authors: | , , |
---|---|
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
eScholarship, University of California
2017
|
Subjects: | |
Online Access: | https://escholarship.org/uc/item/04j576r0 |
id |
ftcdlib:oai:escholarship.org:ark:/13030/qt04j576r0 |
---|---|
record_format |
openpolar |
spelling |
ftcdlib:oai:escholarship.org:ark:/13030/qt04j576r0 2023-05-15T18:01:45+02:00 AD-LIBS: inferring ancestry across hybrid genomes using low-coverage sequence data. Schaefer, Nathan K Shapiro, Beth Green, Richard E 203 2017-04-01 application/pdf https://escholarship.org/uc/item/04j576r0 unknown eScholarship, University of California qt04j576r0 https://escholarship.org/uc/item/04j576r0 public BMC bioinformatics, vol 18, iss 1 Animals Ursidae Humans Markov Chains Hybridization Genetic Polymorphism Single Nucleotide Genome Software Genetic Speciation Admixture Ancestry Bears Speciation Genetics Human Genome Mathematical Sciences Biological Sciences Information and Computing Sciences Bioinformatics article 2017 ftcdlib 2023-02-06T18:41:11Z BackgroundInferring the ancestry of each region of admixed individuals' genomes is useful in studies ranging from disease gene mapping to speciation genetics. Current methods require high-coverage genotype data and phased reference panels, and are therefore inappropriate for many data sets. We present a software application, AD-LIBS, that uses a hidden Markov model to infer ancestry across hybrid genomes without requiring variant calling or phasing. This approach is useful for non-model organisms and in cases of low-coverage data, such as ancient DNA.ResultsWe demonstrate the utility of AD-LIBS with synthetic data. We then use AD-LIBS to infer ancestry in two published data sets: European human genomes with Neanderthal ancestry and brown bear genomes with polar bear ancestry. AD-LIBS correctly infers 87-91% of ancestry in simulations and produces ancestry maps that agree with published results and global ancestry estimates in humans. In brown bears, we find more polar bear ancestry than has been published previously, using both AD-LIBS and an existing software application for local ancestry inference, HAPMIX. We validate AD-LIBS polar bear ancestry maps by recovering a geographic signal within bears that mirrors what is seen in SNP data. Finally, we demonstrate that AD-LIBS is more effective than HAPMIX at inferring ancestry when preexisting phased reference data are unavailable and genomes are sequenced to low coverage.ConclusionsAD-LIBS is an effective tool for ancestry inference that can be used even when few individuals are available for comparison or when genomes are sequenced to low coverage. AD-LIBS is therefore likely to be useful in studies of non-model or ancient organisms that lack large amounts of genomic DNA. AD-LIBS can therefore expand the range of studies in which admixture mapping is a viable tool. Article in Journal/Newspaper polar bear University of California: eScholarship |
institution |
Open Polar |
collection |
University of California: eScholarship |
op_collection_id |
ftcdlib |
language |
unknown |
topic |
Animals Ursidae Humans Markov Chains Hybridization Genetic Polymorphism Single Nucleotide Genome Software Genetic Speciation Admixture Ancestry Bears Speciation Genetics Human Genome Mathematical Sciences Biological Sciences Information and Computing Sciences Bioinformatics |
spellingShingle |
Animals Ursidae Humans Markov Chains Hybridization Genetic Polymorphism Single Nucleotide Genome Software Genetic Speciation Admixture Ancestry Bears Speciation Genetics Human Genome Mathematical Sciences Biological Sciences Information and Computing Sciences Bioinformatics Schaefer, Nathan K Shapiro, Beth Green, Richard E AD-LIBS: inferring ancestry across hybrid genomes using low-coverage sequence data. |
topic_facet |
Animals Ursidae Humans Markov Chains Hybridization Genetic Polymorphism Single Nucleotide Genome Software Genetic Speciation Admixture Ancestry Bears Speciation Genetics Human Genome Mathematical Sciences Biological Sciences Information and Computing Sciences Bioinformatics |
description |
BackgroundInferring the ancestry of each region of admixed individuals' genomes is useful in studies ranging from disease gene mapping to speciation genetics. Current methods require high-coverage genotype data and phased reference panels, and are therefore inappropriate for many data sets. We present a software application, AD-LIBS, that uses a hidden Markov model to infer ancestry across hybrid genomes without requiring variant calling or phasing. This approach is useful for non-model organisms and in cases of low-coverage data, such as ancient DNA.ResultsWe demonstrate the utility of AD-LIBS with synthetic data. We then use AD-LIBS to infer ancestry in two published data sets: European human genomes with Neanderthal ancestry and brown bear genomes with polar bear ancestry. AD-LIBS correctly infers 87-91% of ancestry in simulations and produces ancestry maps that agree with published results and global ancestry estimates in humans. In brown bears, we find more polar bear ancestry than has been published previously, using both AD-LIBS and an existing software application for local ancestry inference, HAPMIX. We validate AD-LIBS polar bear ancestry maps by recovering a geographic signal within bears that mirrors what is seen in SNP data. Finally, we demonstrate that AD-LIBS is more effective than HAPMIX at inferring ancestry when preexisting phased reference data are unavailable and genomes are sequenced to low coverage.ConclusionsAD-LIBS is an effective tool for ancestry inference that can be used even when few individuals are available for comparison or when genomes are sequenced to low coverage. AD-LIBS is therefore likely to be useful in studies of non-model or ancient organisms that lack large amounts of genomic DNA. AD-LIBS can therefore expand the range of studies in which admixture mapping is a viable tool. |
format |
Article in Journal/Newspaper |
author |
Schaefer, Nathan K Shapiro, Beth Green, Richard E |
author_facet |
Schaefer, Nathan K Shapiro, Beth Green, Richard E |
author_sort |
Schaefer, Nathan K |
title |
AD-LIBS: inferring ancestry across hybrid genomes using low-coverage sequence data. |
title_short |
AD-LIBS: inferring ancestry across hybrid genomes using low-coverage sequence data. |
title_full |
AD-LIBS: inferring ancestry across hybrid genomes using low-coverage sequence data. |
title_fullStr |
AD-LIBS: inferring ancestry across hybrid genomes using low-coverage sequence data. |
title_full_unstemmed |
AD-LIBS: inferring ancestry across hybrid genomes using low-coverage sequence data. |
title_sort |
ad-libs: inferring ancestry across hybrid genomes using low-coverage sequence data. |
publisher |
eScholarship, University of California |
publishDate |
2017 |
url |
https://escholarship.org/uc/item/04j576r0 |
op_coverage |
203 |
genre |
polar bear |
genre_facet |
polar bear |
op_source |
BMC bioinformatics, vol 18, iss 1 |
op_relation |
qt04j576r0 https://escholarship.org/uc/item/04j576r0 |
op_rights |
public |
_version_ |
1766171274886774784 |