Bitte benutzen Sie diese Referenz, um auf diese Ressource zu verweisen:
doi:10.22028/D291-42667
Titel: | Chemical representation learning for toxicity prediction |
VerfasserIn: | Born, Jannis Markert, Greta Janakarajan, Nikita Kimber, Talia B. Volkamer, Andrea Martínez, María Rodríguez Manica, Matteo |
Sprache: | Englisch |
Titel: | Digital Discovery |
Bandnummer: | 2 |
Heft: | 3 |
Seiten: | 674-691 |
Verlag/Plattform: | Royal Society of Chemistry |
Erscheinungsjahr: | 2023 |
DDC-Sachgruppe: | 004 Informatik |
Dokumenttyp: | Journalartikel / Zeitschriftenartikel |
Abstract: | Undesired toxicity is a major hindrance to drug discovery and largely responsible for high attrition rates in early stages. This calls for new, reliable, and interpretable molecular property prediction models that help prioritize compounds and thus reduce the high costs for development and the risk to humans, animals, and the environment. Here, we propose an interpretable chemical language model that combines attention with multiscale convolutions and relies on data augmentation. We first benchmark various molecular representations (e.g., fingerprints, different flavors of SMILES and SELFIES, as well as graph and graph kernel methods) revealing that SMILES coupled with augmentation overall yields the best performance. Despite its simplicity, our model is then shown to outperform existing approaches across a wide range of molecular property prediction tasks, including but not limited to toxicity. Moreover, the attention weights of the model allow for easy interpretation and show enrichment of known toxicophores even without explicit supervision. To introduce a notion of model reliability, we propose and combine two simple methods for uncertainty estimation (Monte-Carlo dropout and test-time-augmentation). These methods not only identify samples with high prediction uncertainty, but also allow formation of implicit model ensembles that improve accuracy. Last, we validate our model on a large-scale proprietary toxicity dataset and find that it outperforms previous work while giving similar insights into revealing cytotoxic substructures. |
DOI der Erstveröffentlichung: | 10.1039/D2DD00099G |
URL der Erstveröffentlichung: | https://doi.org/10.1039/D2DD00099G |
Link zu diesem Datensatz: | urn:nbn:de:bsz:291--ds-426678 hdl:20.500.11880/38271 http://dx.doi.org/10.22028/D291-42667 |
ISSN: | 2635-098X |
Datum des Eintrags: | 14-Aug-2024 |
Bezeichnung des in Beziehung stehenden Objekts: | Electronic supplementary information |
In Beziehung stehendes Objekt: | https://www.rsc.org/suppdata/d2/dd/d2dd00099g/d2dd00099g1.pdf |
Fakultät: | MI - Fakultät für Mathematik und Informatik |
Fachrichtung: | MI - Informatik |
Professur: | MI - Prof. Dr. Andrea Volkamer |
Sammlung: | SciDok - Der Wissenschaftsserver der Universität des Saarlandes |
Dateien zu diesem Datensatz:
Datei | Beschreibung | Größe | Format | |
---|---|---|---|---|
d2dd00099g.pdf | 2,36 MB | Adobe PDF | Öffnen/Anzeigen |
Diese Ressource wurde unter folgender Copyright-Bestimmung veröffentlicht: Lizenz von Creative Commons