UNLABELLED: A characterization of the genetic variation of recently admixed populations may reveal historical population events, and is useful for the detection of single nucleotide polymorphisms (SNPs) associated with diseases through association studies and admixture mapping. Inference of locus-specific ancestry is key to our understanding of the genetic variation of such populations. While a number of methods for the inference of locus-specific ancestry are accurate when the ancestral populations are quite distant (e.g. African-Americans), current methods incur a large error rate when inferring the locus-specific ancestry in admixed populations where the ancestral populations are closely related (e.g. Americans of European descent). RESULTS: In this work, we extend previous methods for the inference of locus-specific ancestry by the incorporation of a refined model of recombination events. We present an efficient dynamic programming algorithm to infer the locus-specific ancestries in this model, resulting in a method that attains improved accuracies; the improvement is most significant when the ancestral populations are closely related. An evaluation on a wide range of scenarios, including admixtures of the 52 population groups from the Human Genome Diversity Project demonstrates that locus-specific ancestry can indeed be accurately inferred in these admixtures using our method. Finally, we demonstrate that imputation methods can be improved by the incorporation of locus-specific ancestry, when applied to admixed populations. AVAILABILITY: The implementation of the WINPOP model is available as part of the LAMP package at http://lamp.icsi.berkeley.edu/lamp.

BACKGROUND: Genotype information generated by individual and international efforts carries the promise of revolutionizing disease studies and the association of phenotypes with alleles and haplotypes. Given the enormous amounts of public genotype data, tools for analyzing, interpreting and visualizing these data sets are of critical importance to researchers. In past works we have developed algorithms for genotypes phasing and tag SNP selection, which were shown to be quick and accurate. Both algorithms were available until now only as batch executables. RESULTS: Here we present GEVALT (GEnotype Visualization and ALgorithmic Tool), a software package designed to simplify and expedite the process of genotype analysis, by providing a common interface to several tasks relating to such analysis. GEVALT combines the strong visual abilities of Haploview with our quick and powerful algorithms for genotypes phasing (GERBIL), tag SNP selection (STAMPA) and permutation testing for evaluating significance of association. All of the above are provided in a visually appealing and interactive interface. CONCLUSION: GEVALT is an integrated viewer that uses state of the art phasing and tag SNP selection algorithms. By streamlining the application of GERBIL and STAMPA together with strong visualization for assessment of the results, GEVALT makes the algorithms accessible to the broad community of researchers in genetics.

We present a new algorithm for the problems of genotype phasing and block partitioning. Our algorithm is based on a new stochastic model, and on the novel concept of probabilistic common haplotypes. We formulate the goals of genotype resolving and block partitioning as a maximum likelihood problem, and solve it by an EM algorithm. When applied to real biological SNP data, our algorithm outperforms two state of the art phasing algorithms. Our algorithm is also considerably more sensitive and accurate than a previous method in predicting and identifying disease association.

The study of haplotypes and their diversity in a population is central to disease-association research. We study several problems arising in haplotype block partitioning. Our objective function is the total number of distinct haplotypes in blocks. We show that the problem is NP-hard when there are errors or missing data, and provide approximation algorithms for several of its variants. We also give an algorithm that solves the problem with high probability under a probabilistic model that allows noise and missing data. In addition, we study the multipopulation case, where one has to partition the haplotypes into populations and seek a different block partition in each one. We provide a heuristic for that problem and use it to analyze simulated and real data. On simulated data, our blocks resemble the true partition more than the blocks generated by the LD-based algorithm of Gabriel et al (2002). On single-population real data, we generate a more concise block description than do extant approaches, with better average LD within blocks. The algorithm also gives promising results on real two-population genotype data.