Extracting medicinal chemistry intuition via preference machine learning
Abstract The lead optimization process in drug discovery campaigns is an arduous endeavour where the input of many medicinal chemists is weighed in order to reach a desired molecular property profile. Building the expertise to successfully drive such projects collaboratively is a very time-consuming...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-10-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-023-42242-1 |
| _version_ | 1797558145804926976 |
|---|---|
| author | Oh-Hyeon Choung Riccardo Vianello Marwin Segler Nikolaus Stiefl José Jiménez-Luna |
| author_facet | Oh-Hyeon Choung Riccardo Vianello Marwin Segler Nikolaus Stiefl José Jiménez-Luna |
| author_sort | Oh-Hyeon Choung |
| collection | DOAJ |
| description | Abstract The lead optimization process in drug discovery campaigns is an arduous endeavour where the input of many medicinal chemists is weighed in order to reach a desired molecular property profile. Building the expertise to successfully drive such projects collaboratively is a very time-consuming process that typically spans many years within a chemist’s career. In this work we aim to replicate this process by applying artificial intelligence learning-to-rank techniques on feedback that was obtained from 35 chemists at Novartis over the course of several months. We exemplify the usefulness of the learned proxies in routine tasks such as compound prioritization, motif rationalization, and biased de novo drug design. Annotated response data is provided, and developed models and code made available through a permissive open-source license. |
| first_indexed | 2024-03-10T17:27:20Z |
| format | Article |
| id | doaj.art-5801f4ec45b84ee3be94cf4fc3ca5434 |
| institution | Directory Open Access Journal |
| issn | 2041-1723 |
| language | English |
| last_indexed | 2024-03-10T17:27:20Z |
| publishDate | 2023-10-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj.art-5801f4ec45b84ee3be94cf4fc3ca54342023-11-20T10:08:14ZengNature PortfolioNature Communications2041-17232023-10-0114111010.1038/s41467-023-42242-1Extracting medicinal chemistry intuition via preference machine learningOh-Hyeon Choung0Riccardo Vianello1Marwin Segler2Nikolaus Stiefl3José Jiménez-Luna4Novartis Institutes for Biomedical ResearchNovartis Institutes for Biomedical ResearchMicrosoft Research AI4ScienceNovartis Institutes for Biomedical ResearchMicrosoft Research AI4ScienceAbstract The lead optimization process in drug discovery campaigns is an arduous endeavour where the input of many medicinal chemists is weighed in order to reach a desired molecular property profile. Building the expertise to successfully drive such projects collaboratively is a very time-consuming process that typically spans many years within a chemist’s career. In this work we aim to replicate this process by applying artificial intelligence learning-to-rank techniques on feedback that was obtained from 35 chemists at Novartis over the course of several months. We exemplify the usefulness of the learned proxies in routine tasks such as compound prioritization, motif rationalization, and biased de novo drug design. Annotated response data is provided, and developed models and code made available through a permissive open-source license.https://doi.org/10.1038/s41467-023-42242-1 |
| spellingShingle | Oh-Hyeon Choung Riccardo Vianello Marwin Segler Nikolaus Stiefl José Jiménez-Luna Extracting medicinal chemistry intuition via preference machine learning Nature Communications |
| title | Extracting medicinal chemistry intuition via preference machine learning |
| title_full | Extracting medicinal chemistry intuition via preference machine learning |
| title_fullStr | Extracting medicinal chemistry intuition via preference machine learning |
| title_full_unstemmed | Extracting medicinal chemistry intuition via preference machine learning |
| title_short | Extracting medicinal chemistry intuition via preference machine learning |
| title_sort | extracting medicinal chemistry intuition via preference machine learning |
| url | https://doi.org/10.1038/s41467-023-42242-1 |
| work_keys_str_mv | AT ohhyeonchoung extractingmedicinalchemistryintuitionviapreferencemachinelearning AT riccardovianello extractingmedicinalchemistryintuitionviapreferencemachinelearning AT marwinsegler extractingmedicinalchemistryintuitionviapreferencemachinelearning AT nikolausstiefl extractingmedicinalchemistryintuitionviapreferencemachinelearning AT josejimenezluna extractingmedicinalchemistryintuitionviapreferencemachinelearning |