Journal: Molecular Biology and Evolution

Article Title: Reducing Pervasive False-Positive Identical-by-Descent Segments Detected by Large-Scale Pedigree Analysis

doi: 10.1093/molbev/msu151

Figure Lengend Snippet: Accuracy of child-other IBD segments reported by GERMLINE. ( A ) Heat map of the mean fraction of reported child-other IBD segments contained in a corresponding parent-other segment, binned by two measures of segment length. For each child-other segment, the fraction of the segment also reported as an IBD segment between the parent and the other individual is calculated. Shown in each bin is the mean of the segment fractions calculated for all segments in the bin. ( B ) The fraction of child-other segments that are true IBD as a function of segment length. True IBD segments are defined as having at least 80% of their sites encompassed by a parent-other segment. ( C–F ) Histograms of child-other segment counts binned by segment overlap for segments of 2–3 cM ( C ), 3–4 cM ( D ), 4–5 cM ( E ), and 5–6 cM ( F ). Note the scale changes on the y axes: Though the fraction of true segments of length less than 3 cM is smallest, this range contains roughly 10-fold more true segments than all other length ranges combined.

Article Snippet: Python code implementing HaploScore filtering and the IBD segments analyzed herein are freely available ( https://github.com/23andMe/ibd , last accessed May 8, 2014).

Techniques:

Journal: Molecular Biology and Evolution

Article Title: Reducing Pervasive False-Positive Identical-by-Descent Segments Detected by Large-Scale Pedigree Analysis

doi: 10.1093/molbev/msu151

Figure Lengend Snippet: Graph illustrating the HaploScore computation. The HaploScore for an IBD segment of length L can be represented as the minimum-cost path through the above DAG, where ε denotes the probability of a genotyping error and σ denotes the probability of a switch error at any given site. The DAG has one source, one sink, and one level per genotyped site in the IBD segment. At each level l , the DAG contains four nodes, indicating the haplotype configuration at site l . Each node has weight 0 if the two corresponding alleles are the same, or 1/ ε if they are different. Each node in level l has four outgoing directed edges, one to each of the four nodes in level l + 1. The edge weights are 0, 1/ σ , or 2/ σ , depending on whether 0, 1, or 2 switch errors are necessary to explain the transition. For clarity, some edges are omitted in this figure. The source node src has four outgoing directed edges with weight 0, one to each of the four nodes in level 1. Each node in level L has one outgoing directed edge to the sink node snk with weight 0.

Article Snippet: Python code implementing HaploScore filtering and the IBD segments analyzed herein are freely available ( https://github.com/23andMe/ibd , last accessed May 8, 2014).

Techniques:

Journal: Molecular Biology and Evolution

Article Title: Reducing Pervasive False-Positive Identical-by-Descent Segments Detected by Large-Scale Pedigree Analysis

doi: 10.1093/molbev/msu151

Figure Lengend Snippet: Improving detection of true IBD segments using HaploScore. ( A ) Heat map of the mean fraction of reported IBD segments found in parents, binned by segment genetic length and HaploScore. Calculations are performed as in A . ( B ) Receiver operating characteristic for reported IBD segments of various lengths, discriminating by HaploScore. True IBD is defined as in B . The dashed black line indicates the no-discrimination line. The area under each curve is parenthesized in its legend entry. ( C ) Precision-recall plot for child-other segments binned by segment length.

Article Snippet: Python code implementing HaploScore filtering and the IBD segments analyzed herein are freely available ( https://github.com/23andMe/ibd , last accessed May 8, 2014).

Techniques: