William A. Freyman
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
William A. Freyman has a total of 89 co-authors. Here are the top co-authors by number of shared publications:
- Sebastian Höhna
GeoBio-Center, Ludwig-Maximilians-Universität München, 80333 Munich, Germany
7 shared publications - Carl J. Rothfels
Department of Biology and Ecology Center, Utah State University, Logan, UT 84322 USA
5 shared publications - Bruce G. Baldwin
Jepson Herbarium and Department of Integrative Biology University of California Berkeley 94720‐2465 CA USA
4 shared publications - Adam Auton
23andMe Research Institute, Palo Alto, USA
3 shared publications - Andrew H. Thornhill
N.C.W. Beadle Herbarium, Botany, School of Environmental and Rural Science University of New England Armidale New South Wales Australia
2 shared publications - Bjørn Tore Kopperud
Department of Earth and Environmental Sciences, Paleontology & Geobiology Ludwig‐Maximilians‐Universität München Munich Germany
2 shared publications
Latest Publications
- Open Access
- 2 November 2026
- Published by Wiley in American Journal of Botany
- p. e70284
A tribal‐scale phylogenomic perspective on ancestry and major lineages of the allotetraploid Hawaiian silversword alliance (Madieae, Compositae)
Premise: Disentangling the ancestry of insular plant lineages is important for resolving their evolutionary histories but often complicated by polyploidy, an overrepresented condition in remote island floras. We used a phylogenomic approach to study the origin of subgenomes of the tetraploid Hawaiian silversword alliance and advance understanding of this major example of adaptive radiation. Methods: We phased sequences from 461 target‐capture nuclear loci to subgenomes of the tetraploid silversword alliance (Compositae) using assemblies from diploid North American tarweeds as clade references. We conducted maximum‐likelihood and multispecies‐coalescent analyses of phased loci with and without representatives of diploid lineages throughout tribe Madieae. Results: An allopolyploid origin of the silversword alliance was corroborated, with subgenome A phased consistently to Anisocarpus madioides and subgenome B phased to diploid representatives of each of three other sublineages of the “Madia” lineage. Phylogenomic analyses indicate Anisocarpus madioides is more closely related to subgenome A than to other tarweeds, including Anisocarpus scabridus . Subgenome trees for the silversword alliance are highly congruent with each other and with structural genomic differences between lineages. Conclusions: These results indicate an even closer relationship of the silversword alliance to extant California tarweeds than previously understood. Anisocarpus madioides has chromosomal and ecological traits anticipated in an ancestor of the silversword alliance, with the widest geographic range among perennial tarweeds and sticky bracts completely enveloping fruits. Within the silversword alliance, robust support for four major clades reinforces the phylogenetic value of chromosomal interchanges and taxonomic insights by Asa Gray regarding the enigmatic liana Dubautia latifolia .
- 2 November 2026
- Published by Wiley in American Journal of Botany
- p. e70284
Open AccessA tribal‐scale phylogenomic perspective on ancestry and major lineages of the allotetraploid Hawaiian silversword alliance (Madieae, Compositae)
Premise: Disentangling the ancestry of insular plant lineages is important for resolving their evolutionary histories but often complicated by polyploidy, an overrepresented condition in remote island floras. We used a phylogenomic approach to study the origin of subgenomes of the tetraploid Hawaiian silversword alliance and advance understanding of this major example of adaptive radiation. Methods: We phased sequences from 461 target‐capture nuclear loci to subgenomes of the tetraploid silversword alliance (Compositae) using assemblies from diploid North American tarweeds as clade references. We conducted maximum‐likelihood and multispecies‐coalescent analyses of phased loci with and without representatives of diploid lineages throughout tribe Madieae. Results: An allopolyploid origin of the silversword alliance was corroborated, with subgenome A phased consistently to Anisocarpus madioides and subgenome B phased to diploid representatives of each of three other sublineages of the “Madia” lineage. Phylogenomic analyses indicate Anisocarpus madioides is more closely related to subgenome A than to other tarweeds, including Anisocarpus scabridus . Subgenome trees for the silversword alliance are highly congruent with each other and with structural genomic differences between lineages. Conclusions: These results indicate an even closer relationship of the silversword alliance to extant California tarweeds than previously understood. Anisocarpus madioides has chromosomal and ecological traits anticipated in an ancestor of the silversword alliance, with the widest geographic range among perennial tarweeds and sticky bracts completely enveloping fruits. Within the silversword alliance, robust support for four major clades reinforces the phylogenetic value of chromosomal interchanges and taxonomic insights by Asa Gray regarding the enigmatic liana Dubautia latifolia .
- Open Access
Reconstructing parent genomes using siblings and other relatives
Reconstructing the DNA of ancestors from their descendants has the potential to empower phenotypic analyses (including association and genetic nurture studies), improve pedigree reconstruction, and shed light on the ancestral population and phenotypes of ancestors. We developed HAPI-RECAP, a method that reconstructs the DNA of parents from full siblings and their relatives. This tool leverages HAPI2’s output, a new phasing approach that applies to siblings (and optionally one or both parents) and reliably infers parent haplotypes but does not link the ungenotyped parents’ DNA across chromosomes or between segments flanking ambiguities. By combining IBD between the reconstructed parents and the relatives, HAPI-RECAP resolves the source parent of these segments. Moreover, the method exploits crossovers the children inherited and sex-specific genetic maps to infer the reconstructed parents’ sexes. We validated these methods on research participants from both 23andMe, Inc. and the San Antonio Mexican American Family Studies. Given data for one parent, HAPI2 reconstructs large fractions of the missing parent’s DNA, between 77.6% and 99.97% among all families, and 90.3% on average in three- and four-child families. When reconstructing both parents, HAPI-RECAP inferred between 33.2% and 96.6% of the parents’ genotypes, averaging 70.6% in four-child families. Reconstructed genotypes have average error rates < 10 −3 , or comparable to those from direct genotyping. HAPI-RECAP inferred the parent sexes 100% correctly given IBD-linked segments and can also reconstruct parents without any IBD. As datasets grow in size, more families will be implicitly collected; HAPI-RECAP holds promise to enable high quality parent genotype reconstruction.
Open AccessReconstructing parent genomes using siblings and other relatives
Reconstructing the DNA of ancestors from their descendants has the potential to empower phenotypic analyses (including association and genetic nurture studies), improve pedigree reconstruction, and shed light on the ancestral population and phenotypes of ancestors. We developed HAPI-RECAP, a method that reconstructs the DNA of parents from full siblings and their relatives. This tool leverages HAPI2’s output, a new phasing approach that applies to siblings (and optionally one or both parents) and reliably infers parent haplotypes but does not link the ungenotyped parents’ DNA across chromosomes or between segments flanking ambiguities. By combining IBD between the reconstructed parents and the relatives, HAPI-RECAP resolves the source parent of these segments. Moreover, the method exploits crossovers the children inherited and sex-specific genetic maps to infer the reconstructed parents’ sexes. We validated these methods on research participants from both 23andMe, Inc. and the San Antonio Mexican American Family Studies. Given data for one parent, HAPI2 reconstructs large fractions of the missing parent’s DNA, between 77.6% and 99.97% among all families, and 90.3% on average in three- and four-child families. When reconstructing both parents, HAPI-RECAP inferred between 33.2% and 96.6% of the parents’ genotypes, averaging 70.6% in four-child families. Reconstructed genotypes have average error rates < 10 −3 , or comparable to those from direct genotyping. HAPI-RECAP inferred the parent sexes 100% correctly given IBD-linked segments and can also reconstruct parents without any IBD. As datasets grow in size, more families will be implicitly collected; HAPI-RECAP holds promise to enable high quality parent genotype reconstruction.
- Open Access
- 4 August 2023
- Published by American Association for the Advancement of Science (AAAS) in Science
- Vol. 381 (6657) , 500-+
The genetic legacy of African Americans from Catoctin Furnace
Few African Americans have been able to trace family lineages back to ancestors who died before the 1870 United States Census, the first in which all Black people were listed by name. We analyzed 27 individuals from Maryland’s Catoctin Furnace African American Cemetery (1774–1850), identifying 41,799 genetic relatives among consenting research participants in 23andMe, Inc.’s genetic database. One of the highest concentrations of close relatives is in Maryland, suggesting that descendants of the Catoctin individuals remain in the area. We find that many of the Catoctin individuals derived African ancestry from the Wolof or Kongo groups and European ancestry from Great Britain and Ireland. This study demonstrates the power of joint analysis of historical DNA and large datasets generated through direct-to-consumer ancestry testing.
- 4 August 2023
- Published by American Association for the Advancement of Science (AAAS) in Science
- Vol. 381 (6657) , 500-+
Open AccessThe genetic legacy of African Americans from Catoctin Furnace
Few African Americans have been able to trace family lineages back to ancestors who died before the 1870 United States Census, the first in which all Black people were listed by name. We analyzed 27 individuals from Maryland’s Catoctin Furnace African American Cemetery (1774–1850), identifying 41,799 genetic relatives among consenting research participants in 23andMe, Inc.’s genetic database. One of the highest concentrations of close relatives is in Maryland, suggesting that descendants of the Catoctin individuals remain in the area. We find that many of the Catoctin individuals derived African ancestry from the Wolof or Kongo groups and European ancestry from Great Britain and Ireland. This study demonstrates the power of joint analysis of historical DNA and large datasets generated through direct-to-consumer ancestry testing.
- Open Access
Commonly used Bayesian diversification-rate models produce biologically meaningful differences on empirical phylogenies
Identifying along which lineages shifts in diversification rates occur is a central goal of comparative phylogenetics; these shifts may coincide with key evolutionary events such as the development of novel morphological characters, the acquisition of adaptive traits, polyploidization or other structural genomic changes, or dispersal to a new habitat and subsequent increase in environmental niche space. However, while multiple methods now exist to estimate diversification rates and identify shifts using phylogenetic topologies, the appropriate use and accuracy of these methods is hotly debated. Here we test whether five Bayesian methods—Bayesian Analysis of Macroevolutionary Mixtures ( BAMM ), two implementations of the Lineage-Specific Birth-Death-Shift model ( LSBDS and PESTO ), the approximate Multi-Type Birth-Death model ( MTBD ; implemented in BEAST2 ), and the cladogenetic diversification rate shift model ( CLaDS2 )—produce comparable results. We apply each of these methods to a set of 65 empirical time-calibrated phylogenies and compare inferences of speciation rate, extinction rate, and net diversification rate. We find that the five methods often infer different speciation, extinction, and net-diversification rates. Consequently, these different estimates may lead to different interpretations of the macroevolutionary dynamics. The different estimates can be attributed to fundamental differences among the compared models. Therefore, the inference of shifts in diver-sification rates is strongly method-dependent. We advise biologists to apply multiple methods to test the robustness of the conclusions or to carefully select the method based on the validity of the underlying model assumptions to their particular empirical system. Lay Summary: Understanding why some groups of organisms have more species than others is key to understanding the origin of biodiversity. Theory and empirical evidence suggest that multiple distinct historical events—such as the evolution of particular morphological features (e.g., the flower, the tetrapod limb) and competition amongst species—can produce this pattern of divergent species richness. Identifying when and where on the tree of life shifts in diversification rates occur is important for explaining the origin of modern-day biodiversity and understanding how disparity among species evolves. Several statistical methods have been developed to infer diversification rates and identify these shifts. While these methods each attempt to make inferences about changes in the tempo of diversification, they differ in their underlying statistical models and assumptions. Here we test if these methods draw similar conclusions using a dataset of 65 time-calibrated phylogenies from across multicellular life. We find that inferences of where rate shifts occur strongly depends on the chosen method. Therefore, biologists should choose the model whose assumptions they believe to be the most valid and justify their model choice a priori , or consider using several independent methods to test an evolutionary hypothesis.
Open AccessCommonly used Bayesian diversification-rate models produce biologically meaningful differences on empirical phylogenies
Identifying along which lineages shifts in diversification rates occur is a central goal of comparative phylogenetics; these shifts may coincide with key evolutionary events such as the development of novel morphological characters, the acquisition of adaptive traits, polyploidization or other structural genomic changes, or dispersal to a new habitat and subsequent increase in environmental niche space. However, while multiple methods now exist to estimate diversification rates and identify shifts using phylogenetic topologies, the appropriate use and accuracy of these methods is hotly debated. Here we test whether five Bayesian methods—Bayesian Analysis of Macroevolutionary Mixtures ( BAMM ), two implementations of the Lineage-Specific Birth-Death-Shift model ( LSBDS and PESTO ), the approximate Multi-Type Birth-Death model ( MTBD ; implemented in BEAST2 ), and the cladogenetic diversification rate shift model ( CLaDS2 )—produce comparable results. We apply each of these methods to a set of 65 empirical time-calibrated phylogenies and compare inferences of speciation rate, extinction rate, and net diversification rate. We find that the five methods often infer different speciation, extinction, and net-diversification rates. Consequently, these different estimates may lead to different interpretations of the macroevolutionary dynamics. The different estimates can be attributed to fundamental differences among the compared models. Therefore, the inference of shifts in diver-sification rates is strongly method-dependent. We advise biologists to apply multiple methods to test the robustness of the conclusions or to carefully select the method based on the validity of the underlying model assumptions to their particular empirical system. Lay Summary: Understanding why some groups of organisms have more species than others is key to understanding the origin of biodiversity. Theory and empirical evidence suggest that multiple distinct historical events—such as the evolution of particular morphological features (e.g., the flower, the tetrapod limb) and competition amongst species—can produce this pattern of divergent species richness. Identifying when and where on the tree of life shifts in diversification rates occur is important for explaining the origin of modern-day biodiversity and understanding how disparity among species evolves. Several statistical methods have been developed to infer diversification rates and identify these shifts. While these methods each attempt to make inferences about changes in the tempo of diversification, they differ in their underlying statistical models and assumptions. Here we test if these methods draw similar conclusions using a dataset of 65 time-calibrated phylogenies from across multicellular life. We find that inferences of where rate shifts occur strongly depends on the chosen method. Therefore, biologists should choose the model whose assumptions they believe to be the most valid and justify their model choice a priori , or consider using several independent methods to test an evolutionary hypothesis.
- Open Access
- 13 March 2023
- Published by Wiley in Methods in Ecology and Evolution
- Vol. 14 (5) , 1230-1244
homologizer: Phylogenetic phasing of gene copies into polyploid subgenomes
Organisms such as allopolyploids and F1 hybrids contain multiple distinct subgenomes, each potentially with its own evolutionary history. These organisms present a challenge for multilocus phylogenetic inference and other analyses since it is not apparent which gene copies from different loci are from the same subgenome and thus share an evolutionary history. Here we introduce homologizer, a flexible Bayesian approach that uses a phylogenetic framework to infer the phasing of gene copies across loci into their respective subgenomes. Through the use of simulation tests, we demonstrate that homologizer is robust to a wide range of factors, such as incomplete lineage sorting and the phylogenetic informativeness of loci. Furthermore, we establish the utility of homologizer on real data, by analysing a multilocus dataset consisting of nine diploids and 19 tetraploids from the fern family Cystopteridaceae. Finally, we describe how homologizer may potentially be used beyond its core phasing functionality to identify non‐homologous sequences, such as hidden paralogs or contaminants.
- 13 March 2023
- Published by Wiley in Methods in Ecology and Evolution
- Vol. 14 (5) , 1230-1244
Open Accesshomologizer: Phylogenetic phasing of gene copies into polyploid subgenomes
Organisms such as allopolyploids and F1 hybrids contain multiple distinct subgenomes, each potentially with its own evolutionary history. These organisms present a challenge for multilocus phylogenetic inference and other analyses since it is not apparent which gene copies from different loci are from the same subgenome and thus share an evolutionary history. Here we introduce homologizer, a flexible Bayesian approach that uses a phylogenetic framework to infer the phasing of gene copies across loci into their respective subgenomes. Through the use of simulation tests, we demonstrate that homologizer is robust to a wide range of factors, such as incomplete lineage sorting and the phylogenetic informativeness of loci. Furthermore, we establish the utility of homologizer on real data, by analysing a multilocus dataset consisting of nine diploids and 19 tetraploids from the fern family Cystopteridaceae. Finally, we describe how homologizer may potentially be used beyond its core phasing functionality to identify non‐homologous sequences, such as hidden paralogs or contaminants.
- Open Access
adequaSSE: Model Adequacy Testing for Trait-Dependent Diversification Models
The presence of large variation in speciation and extinctin rates across the tree of life has long been hypothesized to be driven by the evolution of traits that affect diversification. To test such hypotheses, phylogenetic biologists have developed a wide class of state-dependent birth-death processes that jointly model trait evolution and the diversification process. However, it has since been shown that these models are sensitive to falsely linking traits to diversification. Here we present a Bayesian approach to test the adequacy of statedependent birth-death models by statistically checking whether they describe the variation observed in the data. Our method generates posterior predictive distributions for a suite of informative test statistics, providing a general framework for testing diversification processes and models of trait evolution. We simulate data sets under different violations of model assumptions and find that our approach successfully detects the inadequacy of the model for them. We further show that the manner in which a model fails to fit the data can reveal insights into the processes of trait evolution and diversification.
Open AccessadequaSSE: Model Adequacy Testing for Trait-Dependent Diversification Models
The presence of large variation in speciation and extinctin rates across the tree of life has long been hypothesized to be driven by the evolution of traits that affect diversification. To test such hypotheses, phylogenetic biologists have developed a wide class of state-dependent birth-death processes that jointly model trait evolution and the diversification process. However, it has since been shown that these models are sensitive to falsely linking traits to diversification. Here we present a Bayesian approach to test the adequacy of statedependent birth-death models by statistically checking whether they describe the variation observed in the data. Our method generates posterior predictive distributions for a suite of informative test statistics, providing a general framework for testing diversification processes and models of trait evolution. We simulate data sets under different violations of model assumptions and find that our approach successfully detects the inadequacy of the model for them. We further show that the manner in which a model fails to fit the data can reveal insights into the processes of trait evolution and diversification.
- Open Access
- 4 November 2021
- Published by Elsevier in The American Journal of Human Genetics
- Vol. 108 (11) , 2052-2070
Bonsai: An efficient method for inferring large human pedigrees from genotype data
- 4 November 2021
- Published by Elsevier in The American Journal of Human Genetics
- Vol. 108 (11) , 2052-2070
Open AccessBonsai: An efficient method for inferring large human pedigrees from genotype data
- 30 October 2021
- Published by Wiley in Methods in Ecology and Evolution
- Vol. 13 (2) , 314-323
RevGadgets: An R package for visualizing Bayesian phylogenetic analyses from RevBayes
Statistical phylogenetic methods are the foundation for a wide range of evolutionary and epidemiological studies. However, as these methods grow increasingly complex, users often encounter significant challenges with summarizing, visualizing and communicating their key results. We presentRevGadgets, anRpackage for creating publication‐quality figures from the results of a large variety of phylogenetic analyses performed inRevBayes(and other phylogenetic software packages). We demonstrate how to useRevGadgetsthrough a set of vignettes that cover the most common use cases that researchers will encounter. RevGadgetsis an open‐source, extensible package that will continue to evolve in parallel withRevBayes, helping researchers to make sense of and communicate the results of a diverse array of analyses.
- 30 October 2021
- Published by Wiley in Methods in Ecology and Evolution
- Vol. 13 (2) , 314-323
RevGadgets: An R package for visualizing Bayesian phylogenetic analyses from RevBayes
Statistical phylogenetic methods are the foundation for a wide range of evolutionary and epidemiological studies. However, as these methods grow increasingly complex, users often encounter significant challenges with summarizing, visualizing and communicating their key results. We presentRevGadgets, anRpackage for creating publication‐quality figures from the results of a large variety of phylogenetic analyses performed inRevBayes(and other phylogenetic software packages). We demonstrate how to useRevGadgetsthrough a set of vignettes that cover the most common use cases that researchers will encounter. RevGadgetsis an open‐source, extensible package that will continue to evolve in parallel withRevBayes, helping researchers to make sense of and communicate the results of a diverse array of analyses.
- 25 October 2021
- Published by Wiley in American Journal of Botany
- Vol. 108 (10) , 2015-2037
Directionally biased habitat shifts and biogeographically informative cytonuclear discordance in the Hawaiian silversword alliance (Compositae)
Premise: Expanded phylogenetic analyses of the Hawaiian silversword alliance (Argyroxiphium, Dubautia, Wilkesia; Compositae) were undertaken to assess evolutionary and biogeographic informativeness of cytonuclear discordance and any biases in evolutionary directionality of ecological transitions within this prominent example of adaptive radiation.Methods: Samples spanning the geographic and ecological distributions of all recognized taxa were included in phylogenetic and biogeographic analyses of nuclear ribosomal DNA (nrDNA) and cpDNA sequences. Bayesian model testing approaches were used to model ecological evolution and the evolution of nuclear chromosomal arrangements while accounting for phylogenetic uncertainty.Results: Cytonuclear discordance detected previously appears to reflect chloroplast capture, at least in part, with nrDNA trees being largely congruent with nuclear chromosomal structural data and fine‐scale taxonomy. Comparison of biogeographic histories estimated from the posterior distributions of nrDNA and cpDNA trees, including inferred chloroplast‐capture events, provides additional resolution of dispersal history, including a back‐dispersal to Maui Nui from Hawai‘i. A newly resolved major nrDNA clade of endemic Kaua‘i taxa that mostly were described as new‐to‐science since the 1980s strengthens the earlier hypothesis that diversification on Kaua‘i has not waned since the island began to decline in area through subsidence and erosion. Bias in habitat shifts was estimated, with transitions from dry‐to‐mesic or ‐wet and from wet‐to‐mesic or ‐bog habitats dominating diversification of the silversword alliance from a dry‐adapted tarweed ancestor.Conclusions: The habitat‐transition biases estimated here may indicate more limited pathways of ecological evolution than proposed previously for an adaptive radiation involving such major ecological shifts.
- 25 October 2021
- Published by Wiley in American Journal of Botany
- Vol. 108 (10) , 2015-2037
Directionally biased habitat shifts and biogeographically informative cytonuclear discordance in the Hawaiian silversword alliance (Compositae)
Premise: Expanded phylogenetic analyses of the Hawaiian silversword alliance (Argyroxiphium, Dubautia, Wilkesia; Compositae) were undertaken to assess evolutionary and biogeographic informativeness of cytonuclear discordance and any biases in evolutionary directionality of ecological transitions within this prominent example of adaptive radiation.Methods: Samples spanning the geographic and ecological distributions of all recognized taxa were included in phylogenetic and biogeographic analyses of nuclear ribosomal DNA (nrDNA) and cpDNA sequences. Bayesian model testing approaches were used to model ecological evolution and the evolution of nuclear chromosomal arrangements while accounting for phylogenetic uncertainty.Results: Cytonuclear discordance detected previously appears to reflect chloroplast capture, at least in part, with nrDNA trees being largely congruent with nuclear chromosomal structural data and fine‐scale taxonomy. Comparison of biogeographic histories estimated from the posterior distributions of nrDNA and cpDNA trees, including inferred chloroplast‐capture events, provides additional resolution of dispersal history, including a back‐dispersal to Maui Nui from Hawai‘i. A newly resolved major nrDNA clade of endemic Kaua‘i taxa that mostly were described as new‐to‐science since the 1980s strengthens the earlier hypothesis that diversification on Kaua‘i has not waned since the island began to decline in area through subsidence and erosion. Bias in habitat shifts was estimated, with transitions from dry‐to‐mesic or ‐wet and from wet‐to‐mesic or ‐bog habitats dominating diversification of the silversword alliance from a dry‐adapted tarweed ancestor.Conclusions: The habitat‐transition biases estimated here may indicate more limited pathways of ecological evolution than proposed previously for an adaptive radiation involving such major ecological shifts.
- Open Access
- 23 December 2020
- Published by Oxford University Press (OUP) in Molecular Biology and Evolution
- Vol. 38 (5) , 2131-2151
Fast and Robust Identity-by-Descent Inference with the Templated Positional Burrows–Wheeler Transform
Estimating the genomic location and length of identical-by-descent (IBD) segments among individuals is a crucial step in many genetic analyses. However, the exponential growth in the size of biobank and direct-to-consumer genetic data sets makes accurate IBD inference a significant computational challenge. Here we present the templated positional Burrows–Wheeler transform (TPBWT) to make fast IBD estimates robust to genotype and phasing errors. Using haplotype data simulated over pedigrees with realistic genotyping and phasing errors, we show that the TPBWT outperforms other state-of-the-art IBD inference algorithms in terms of speed and accuracy. For each phase-aware method, we explore the false positive and false negative rates of inferring IBD by segment length and characterize the types of error commonly found. Our results highlight the fragility of most phased IBD inference methods; the accuracy of IBD estimates can be highly sensitive to the quality of haplotype phasing. Additionally, we compare the performance of the TPBWT against a widely used phase-free IBD inference approach that is robust to phasing errors. We introduce both in-sample and out-of-sample TPBWT-based IBD inference algorithms and demonstrate their computational efficiency on massive-scale data sets with millions of samples. Furthermore, we describe the binary file format for TPBWT-compressed haplotypes that results in fast and efficient out-of-sample IBD computes against very large cohort panels. Finally, we demonstrate the utility of the TPBWT in a brief empirical analysis, exploring geographic patterns of haplotype sharing within Mexico. Hierarchical clustering of IBD shared across regions within Mexico reveals geographically structured haplotype sharing and a strong signal of isolation by distance. Our software implementation of the TPBWT is freely available for noncommercial use in the code repository (https://github.com/23andMe/phasedibd, last accessed January 11, 2021).
- 23 December 2020
- Published by Oxford University Press (OUP) in Molecular Biology and Evolution
- Vol. 38 (5) , 2131-2151
Open AccessFast and Robust Identity-by-Descent Inference with the Templated Positional Burrows–Wheeler Transform
Estimating the genomic location and length of identical-by-descent (IBD) segments among individuals is a crucial step in many genetic analyses. However, the exponential growth in the size of biobank and direct-to-consumer genetic data sets makes accurate IBD inference a significant computational challenge. Here we present the templated positional Burrows–Wheeler transform (TPBWT) to make fast IBD estimates robust to genotype and phasing errors. Using haplotype data simulated over pedigrees with realistic genotyping and phasing errors, we show that the TPBWT outperforms other state-of-the-art IBD inference algorithms in terms of speed and accuracy. For each phase-aware method, we explore the false positive and false negative rates of inferring IBD by segment length and characterize the types of error commonly found. Our results highlight the fragility of most phased IBD inference methods; the accuracy of IBD estimates can be highly sensitive to the quality of haplotype phasing. Additionally, we compare the performance of the TPBWT against a widely used phase-free IBD inference approach that is robust to phasing errors. We introduce both in-sample and out-of-sample TPBWT-based IBD inference algorithms and demonstrate their computational efficiency on massive-scale data sets with millions of samples. Furthermore, we describe the binary file format for TPBWT-compressed haplotypes that results in fast and efficient out-of-sample IBD computes against very large cohort panels. Finally, we demonstrate the utility of the TPBWT in a brief empirical analysis, exploring geographic patterns of haplotype sharing within Mexico. Hierarchical clustering of IBD shared across regions within Mexico reveals geographically structured haplotype sharing and a strong signal of isolation by distance. Our software implementation of the TPBWT is freely available for noncommercial use in the code repository (https://github.com/23andMe/phasedibd, last accessed January 11, 2021).