AUTHOR=Jiang Jue , Wei Xuxu , Lu YuKang , Li Simin , Xu Xue TITLE=Network-based prediction of anti-cancer drug combinations JOURNAL=Frontiers in Pharmacology VOLUME=Volume 15 - 2024 YEAR=2024 URL=https://www.frontiersin.org/journals/pharmacology/articles/10.3389/fphar.2024.1418902 DOI=10.3389/fphar.2024.1418902 ISSN=1663-9812 ABSTRACT=Drug combinations have emerged as a promising therapeutic approach in cancer treatment, aimed at overcoming drug resistance and improving the efficacy of monotherapy regimens. However, identifying effective drug combinations has traditionally been time-consuming and often dependent on chance discoveries. Therefore, there is an urgent need to explore alternative strategies to support experimental research. In this study, we propose network-based prediction models to identify potential drug combinations for 11 types of cancer. Our approach involves extracting 55,299 associations from literature and constructing human protein interactomes for each cancer type. To predict drug combinations, we measure the proximity of drug-drugData available on request from the corresponding authors.We first collected cancer-related literatures dated from 1995 to the 2020 in PubMed.Keywords including drug, gene, cancer cell line, cancer type and drug treatment outcome were extracted from each abstract, and sematic associations among 6,261 drugs, 3,764 genes, 2,002 cancer cell lines and 73 cancer types were identified. After standardization of drugs, genes, cancer cell lines and cancer types based on PubChem, DrugBank and UniProt, associations between 1317 drugs and 1,315 genes were obtained. Next, background protein-protein interactions (PPIs) were assembled from 5 data sources, i.e., BioGRID (https://thebiogrid.org/), MINT (https://mint.bio.uniroma2.it/), BIND (http://bind.ca), DIP (http://dip.doe-mbi.ucla.edu), IntAct (https://www.ebi.ac.uk/intact/) and HPRD (http://hprd.org/index_html), which included 27,123 nodes and 663,114 edges. Based on the PubMed-derived drug-gene associations and the background network, we constructed networks for 11 most common cancer subtypes, including non-small cell lung cancer (NSCLC), colon cancer, acute myeloid leukemia (AML), pancreatic cancer, breast cancer, ovarian cancer, liver cancer, prostate cancer, osteosarcoma, glioma (Fig.