| Popis: |
Motivation: Hepatocellular carcinoma (HCC) is a significant health problem worldwide and annual number of cases are nearly more than 700,000. However,there are few safe and effective thera-peutic options for HCC patients.Here, we propose a new approach for predicting therapeutic and aggravating drugs for HCCbased ontissue-specificpathways, which considersnot onlyliver tis-sueand functional informationof pathways, but also the changes of single gene in pathways. Results: Firstly, we map genes related to HCC to the liver-specific protein interaction network and get anextended tissue-specific gene set of HCC. Then, based on the extended gene set, 12 en-riched KEGGfunctional pathways are extracted.Using Kolmogorov-Smirnov statistic, we calculate the therapeutic scores of drugs based on the 12 tissue-specific pathways. Finally, after filtering by Comparative Toxicogenomics Database (CTD) benchmark,we get 3 therapeutic drugs and 3 ag-gravating drugs for HCC. Furthermore, we validate the6potentially related drugs of HCC by analyz-ingtheiroverlaps with drug indications reported in PubMed literatures, and also making CMapprofile similarity analysis and KEGG enrichment analysis based on their targets. All analysis results sug-gest thatour approach is effective and accurate for discovering novel therapeutic options for HCC and it can be easily extended to other diseases.More importantly, our method can clearly distin-guish therapeutic and aggravating drugsforHCC, which also can be used to indicateunmarked drug-disease associations in CTD as positive or negative. |
