Recent advances in sequencing technology is enabling fast and cost-effective generation of sequence data. Soon, whole-genome sequencing will become a routine assay, opening up new opportunities for biomedical research and related fields. The overall objective of this proposal is to develop accurate, scalable computational methods for whole-genome variation study, both by developing a new theoretical framework for studying population genetics models and by improving the key mathematical component underlying various statistical tools for genome analysis. A common thread that runs through this proposal is recombination. Both crossover and gene conversion recombinations will be considered.
The specific aims of the proposed research are: (1) Develop and apply a new theoretical framework that complements the standard coalescent theory when the rate of recombination is moderate to large, thus providing a new set of useful analytic tools to the population genomics community. (2) Devise principled methods to derive accurate multi-locus conditional sampling distributions directly from the underlying population genetics model. Improve the accuracy of a wide range of statistical methods for genome analysis that utilize conditional sampling distributions. (3) Develop scalable computational methods for joint estimation of crossover rates, gene conversion rates, and mean conversion tract lengths from population SNP data. Incorporate realistic biological scenarios into a model with overlapping gene conversions and extend the model to handle ectopic (or non-allelic) gene conversions in multigene families. The above goals will be achieved not by engineering modifications based on intuition or simulations, but by applying and generalizing recent mathematical results that are rigorous and accurate. The new theoretical framework developed in this research will allow one to carry out analytic computation, which was considered to be intractable in the standard coalescent theory with recombination. Furthermore, mathematically justified approximations based on diffusion processes will be devised and portable software packages will be developed for population genomics analysis.

Public Health Relevance

The ongoing large-scale sequencing projects will provide a comprehensive view of genomic variation in populations, helping to unravel the genetic basis of human biology and disease risk. Recombination is a major biological mechanism responsible for generating genetic variation in a population, and has important implications for many computational problems in genome analysis, including disease-association mapping and detecting signatures of natural selection. The proposed research will help with analyzing and interpreting whole-genome variation data, by developing novel mathematical frameworks and statistical tools for studying population genetics models with recombination.

National Institute of Health (NIH)
National Institute of General Medical Sciences (NIGMS)
Research Project (R01)
Project #
Application #
Study Section
Genetic Variation and Evolution Study Section (GVE)
Program Officer
Eckstrand, Irene A
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of California Berkeley
Engineering (All Types)
Schools of Engineering
United States
Zip Code
Miroshnikov, Alexey; Steinrücken, Matthias (2017) Computing the joint distribution of the total tree length across loci in populations with variable size. Theor Popul Biol 118:1-19
Terhorst, Jonathan; Kamm, John A; Song, Yun S (2017) Robust and scalable inference of population history from hundreds of unphased whole genomes. Nat Genet 49:303-309
Crawford, Nicholas G; Kelly, Derek E; Hansen, Matthew E B et al. (2017) Loci associated with skin pigmentation identified in African populations. Science 358:
Luo, Shishi; Mattingly, Jonathan C (2017) SCALING LIMITS OF A MODEL FOR SELECTION AT TWO SCALES. Nonlinearity 30:1682-1707
Mallick, Swapan; Li, Heng; Lipson, Mark et al. (2016) The Simons Genome Diversity Project: 300 genomes from 142 diverse populations. Nature 538:201-206
Jewett, Ethan M; Steinrücken, Matthias; Song, Yun S (2016) The Effects of Population Size Histories on Estimates of Selection Coefficients from Time-Series Genetic Data. Mol Biol Evol 33:3002-3027
Steinrücken, Matthias; Jewett, Ethan M; Song, Yun S (2016) SpectralTDF: transition densities of diffusion processes with time-varying selection parameters, mutation rates and effective population sizes. Bioinformatics 32:795-7
Luo, Shishi; Yu, Jane A; Song, Yun S (2016) Estimating Copy Number and Allelic Variation at the Immunoglobulin Heavy Chain Locus Using Short Reads. PLoS Comput Biol 12:e1005117
Sheehan, Sara; Song, Yun S (2016) Deep Learning for Population Genetic Inference. PLoS Comput Biol 12:e1004845
Zou, James Y; Park, Danny S; Burchard, Esteban G et al. (2015) Genetic and socioeconomic study of mate choice in Latinos reveals novel assortment patterns. Proc Natl Acad Sci U S A 112:13621-6

Showing the most recent 10 out of 34 publications