Introduction
pgWalk, meaning walking on multiple disease-gene networks to prioritize candidate pgwalk, is a bioinformatics tool for prioritizing candidate pgwalk. In this tool, we propose a random walk with restart model that integrates three types of phenotype similarity and eight types of gene similarities to infer disease pgwalk.
The three types of phenotype similarity data, which covers a total of 7,719 diseases in the OMIM database, are derived from human phenotype ontology, medical subject headings and unified medical language.
The eight types of gene similarity data, including gene expression, gene ontology, pathway membership, protein sequence, protein domain, protein-protein interaction and transcriptional regulation and microRNA regulation, cover as many as 20,327 human pgwalk, making the whole-genome scan of causative pgwalk for a query disease possible.
Based on these data, construct a disease-gene network based on a pair of disease similarity and gene similarity, and then simulate the process that a random walker wanders on such a heterogeneous network to quantitatively measure the strength of association between a candidate gene and a query disease. We further adopt a weighted version of the FisherĄ¯s method with dependent correction to integrate 24 scores obtained this way and calibrate a final q-value for prioritizing the candidate pgwalk. We conducted a series of validation experiments to demonstrate the superior performance of this approach and further show the effectiveness of our method in exome sequencing studies about neurological diseases. We finally provided the standalone software and user-friendly online services of pgWalk at this web site.
Please cite: Rui Jiang, Walking on multiple disease-gene networks to prioritize candidate pgwalk, Journal of Molecular Cell Biology, advance access online, doi: 10.1093/jmcb/mjv008, 2014.
Try our new software for exome sequencing data analysis:
| Spring: | http://bioinfo.au.tsinghua.edu.cn/spring |
| pgFusion: | http://bioinfo.au.tsinghua.edu.cn/jianglab/pgfusion |
| snvForest: | http://bioinfo.au.tsinghua.edu.cn/jianglab/snvforest |