For more than a decade, the ChEMBL database has grown to incorporate ever more data, extracted from the scientific literature or directly deposited. Initially focused on small molecules bioactivities, it has evolved to fulfil (and indeed, sometimes to exceed) the expectations of the scientific community in the challenging field of drug discovery. Now also providing drug data such as development status, mechanisms of action, indications and withdrawal reasons, ChEMBL supports more than ever pharmaceutical research. The database comprises in its latest version [1] 1.1 million assays for 1.9 million compounds and more than 8,000 protein targets, with 26,000 indications and 4700 mechanisms. All of this, freely accessible to the global community. This information can be used as a source of evidence but many scientists also find in<br/>ChEMBL the data they need to build structure activity relationship models [2] . With the excitement surrounding machine learning and AI, ChEMBL has a central role to play, addressing the need for more and more quality data. Through this presentation, I will describe several important details about the ChEMBL database and how its data are used to develop predictive models.<br/>

ChEMBL: Leading data source for machine learning applied to drug discovery

Macromolecular Chemistry: The Second Century

[Author Benefits](https://scimeetings.acs.org/?utm_source=pubs_content_marketing&utm_medium=website&utm_campaign=0320_MCF_NPI_Launch_Spring_Homepage&ref=pubs_content_marketing)

[How to use SciMeetings](https://storage.googleapis.com/dl.morressier.com/acs/SciMeetings-ACSSpring2020-HowToGuide_M3020-Final.pdf)

[Frequently Asked Questions](https://storage.googleapis.com/dl.morressier.com/acs/SciMtng_ACSNMSp2020_FAQ_M3120_Final.pdf)

ACS Spring 2020 National Meeting & Expo

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.