{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Parrilla-Gutierrez JM"],"funding":["RCUK | Engineering and Physical Sciences Research Council (EPSRC)","European Research Council","RCUK | Engineering and Physical Sciences Research Council"],"pagination":["200-209"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC10965440"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["4(3)"],"pubmed_abstract":["Here we present a machine learning model trained on electron density for the production of host-guest binders. These are read out as simplified molecular-input line-entry system (SMILES) format with >98% accuracy, enabling a complete characterization of the molecules in two dimensions. Our model generates three-dimensional representations of the electron density and electrostatic potentials of host-guest systems using a variational autoencoder, and then utilizes these representations to optimize the generation of guests via gradient descent. Finally the guests are converted to SMILES using a transformer. The successful practical application of our model to established molecular host systems, cucurbit[n]uril and metal-organic cages, resulted in the discovery of 9 previously validated guests for CB[6] and 7 unreported guests (with association constant K<sub>a</sub> ranging from 13.5 M<sup>-1</sup> to 5,470 M<sup>-1</sup>) and the discovery of 4 unreported guests for [Pd<sub>2</sub>1<sub>4</sub>]<sup>4+</sup> (with K<sub>a</sub> ranging from 44 M<sup>-1</sup> to 529 M<sup>-1</sup>)."],"journal":["Nature computational science"],"pubmed_title":["Electron density-based GPT for optimization and suggestion of host-guest binders."],"pmcid":["PMC10965440"],"funding_grant_id":["EP/L023652/1, EP/R020914/1, EP/S030603/1, EP/R01308X/1, EP/S017046/1, and EP/S019472/1","670467"],"pubmed_authors":["Ayme JF","Cronin L","Wilbraham L","Parrilla-Gutierrez JM","Granda JM","Bajczyk MD"],"additional_accession":[]},"is_claimable":false,"name":"Electron density-based GPT for optimization and suggestion of host-guest binders.","description":"Here we present a machine learning model trained on electron density for the production of host-guest binders. These are read out as simplified molecular-input line-entry system (SMILES) format with >98% accuracy, enabling a complete characterization of the molecules in two dimensions. Our model generates three-dimensional representations of the electron density and electrostatic potentials of host-guest systems using a variational autoencoder, and then utilizes these representations to optimize the generation of guests via gradient descent. Finally the guests are converted to SMILES using a transformer. The successful practical application of our model to established molecular host systems, cucurbit[n]uril and metal-organic cages, resulted in the discovery of 9 previously validated guests for CB[6] and 7 unreported guests (with association constant K<sub>a</sub> ranging from 13.5 M<sup>-1</sup> to 5,470 M<sup>-1</sup>) and the discovery of 4 unreported guests for [Pd<sub>2</sub>1<sub>4</sub>]<sup>4+</sup> (with K<sub>a</sub> ranging from 44 M<sup>-1</sup> to 529 M<sup>-1</sup>).","dates":{"release":"2024-01-01T00:00:00Z","publication":"2024 Mar","modification":"2025-04-27T03:11:25.513Z","creation":"2025-04-06T18:45:09.016Z"},"accession":"S-EPMC10965440","cross_references":{"pubmed":["38459272"],"doi":["10.1038/s43588-024-00602-x"]}}