The Atomwise Library
In April 2019 Chris Swain and Mat Todd, on behalf of the Open Source Antibiotics Mur Ligase group, applied for an AIMS Awards program grant from Atomwise. The proposal is here. The award was made in June 2019 (project A19-229). The grant was for "i) a customized small molecule virtual screen using Atomwise's AI technology, ii) ~72 compounds predicted to bind to your specific target protein - shipped to you ready for testing, and iii) support from Atomwise's medicinal and computational chemists."
Despite all parties agreeing to the open source methodology of the project, there were some extensive delays over contract negotiations. The compounds were designed by October 2020 and (96) samples were shipped to UCL in Nov 2020. There were then further contract delays allowing the compounds to be shipped from UCL to the University of Warwick and the Diamond Light Source. An Excel spreadsheet and sdf file containing SMILES and other data for all compounds are available for download. Images of all structures may be found in Issue #33.
The compounds were screened against EcMurE using the XChem platform by Laura Diaz Saez in October 2021, identifying 8 binders. Follow-up SPR was carried out by Laura in November 2021 with the files posted here. Evaluation of the ability of the compounds to inhibit several mur ligases was carried out by Adrian Lloyd and team at Warwick University in Sept 2022, with data posted here.
