Machine learning for chemical discovery
Discovering chemicals with desired attributes is a long and painstaking process. Curated datasets containing reliable quantum-mechanical properties for millions of molecules are becoming increasingly available. The development of novel machine learning tools to obtain chemical knowledge from these d...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2020-08-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-020-17844-8 |
| _version_ | 1819198670165245952 |
|---|---|
| author | Alexandre Tkatchenko |
| author_facet | Alexandre Tkatchenko |
| author_sort | Alexandre Tkatchenko |
| collection | DOAJ |
| description | Discovering chemicals with desired attributes is a long and painstaking process. Curated datasets containing reliable quantum-mechanical properties for millions of molecules are becoming increasingly available. The development of novel machine learning tools to obtain chemical knowledge from these datasets has the potential to revolutionize the process of chemical discovery. Here, I comment on recent breakthroughs in this emerging field and discuss the challenges for the years to come. |
| first_indexed | 2024-12-23T03:04:08Z |
| format | Article |
| id | doaj.art-316c9b03a1074f64b5fbab43c739f1f8 |
| institution | Directory Open Access Journal |
| issn | 2041-1723 |
| language | English |
| last_indexed | 2024-12-23T03:04:08Z |
| publishDate | 2020-08-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj.art-316c9b03a1074f64b5fbab43c739f1f82022-12-21T18:02:21ZengNature PortfolioNature Communications2041-17232020-08-011111410.1038/s41467-020-17844-8Machine learning for chemical discoveryAlexandre Tkatchenko0Department of Physics and Materials Science, University of LuxembourgDiscovering chemicals with desired attributes is a long and painstaking process. Curated datasets containing reliable quantum-mechanical properties for millions of molecules are becoming increasingly available. The development of novel machine learning tools to obtain chemical knowledge from these datasets has the potential to revolutionize the process of chemical discovery. Here, I comment on recent breakthroughs in this emerging field and discuss the challenges for the years to come.https://doi.org/10.1038/s41467-020-17844-8 |
| spellingShingle | Alexandre Tkatchenko Machine learning for chemical discovery Nature Communications |
| title | Machine learning for chemical discovery |
| title_full | Machine learning for chemical discovery |
| title_fullStr | Machine learning for chemical discovery |
| title_full_unstemmed | Machine learning for chemical discovery |
| title_short | Machine learning for chemical discovery |
| title_sort | machine learning for chemical discovery |
| url | https://doi.org/10.1038/s41467-020-17844-8 |
| work_keys_str_mv | AT alexandretkatchenko machinelearningforchemicaldiscovery |