Skip to main content
Fig. 1 | BMC Bioinformatics

Fig. 1

From: SVhound: detection of regions that harbor yet undetected structural variation

Fig. 1

Overview and evaluation of SVhound based on 1000 genomes data set. A Computing the probabilities of detecting new SV-alleles in a window. First, the chromosome is divided into non overlapping windows. For each window the number of distinct observed SV-alleles is counted and the diversity parameter is estimated Eq. 2 (see “Methods”). Finally, the probability of detecting a clairvoyant SV (clSV) (\(p_{new}\)) for each particular window is computed using Eq. 3 (see “Methods”). B Scatterplots showing predictive power (correlation) between \(p_{new}\) and the fraction of undetected SV for a 10kbp and 100kbp window and two sample sizes 100 genomes (top panels) and 1000 genomes (bottom panels), sub-sampled from the 1KGP data. The x-axis shows the prediction made by SVhound (probability of new SV-allele, \(p_{new}\)) and the y-axis shows the proportion of undetected SV-alleles in the non-sampled individuals (\(f_{undetected}\)). Be aware that the axis ranges have been adopted to better visualize the results. Note that regardless of sample size, SVhound performs better in the 100kbp window when comparing both window lengths. C Distribution of the probabilities of detecting a clSV (\(p_{new}\)) for different window lengths

Back to article page