Fig. 2
From: Fifteen years of ChEMBL and its role in cheminformatics and drug discovery
Wordcloud produced by using the PubMed API to retrieve articles between 2010 and 2024 (912 papers in total) that mention “ChEMBL” in either the title or abstract (using the nlkt and wordcloud packages in python; number of common keywords extracted = 30). These are the most common keywords identified: ('drug', 1105), ('molecular', 824), ('models', 812), ('targets', 635), ('molecules', 606), ('activity', 591), ('inhibitors', 590), ('target', 565), ('screening', 536), ('drugs', 522), ('potential', 495), ('learning', 476), ('binding', 440), ('prediction', 433), ('protein', 390), ('machine', 377), ('docking', 360), ('novel', 348), ('virtual', 330), ('design', 297), ('methods', 291), ('active', 259), ('space', 246), ('structural', 243), ('biological', 242), ('qsar', 238), ('structures', 237), ('ligands', 234), ('interactions', 230), ('performance', 223)