Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
hERG-toxicity prediction using traditional machine learning and advanced deep learning techniques
RISE Research Institutes of Sweden, Digital Systems, Data Science.
RISE Research Institutes of Sweden, Bioeconomy and Health, Chemical and Pharmaceutical Toxicology.ORCID iD: 0000-0003-4158-4148
RISE Research Institutes of Sweden, Digital Systems, Data Science.
RISE Research Institutes of Sweden, Digital Systems, Data Science.
Show others and affiliations
2023 (English)In: Current Research in Toxicology, ISSN 2666-027X, Vol. 5, article id 100121Article in journal (Refereed) Published
Abstract [en]

The rise of artificial intelligence (AI) based algorithms has gained a lot of interest in the pharmaceutical development field. Our study demonstrates utilization of traditional machine learning techniques such as random forest (RF), support-vector machine (SVM), extreme gradient boosting (XGBoost), deep neural network (DNN) as well as advanced deep learning techniques like gated recurrent unit-based DNN (GRU-DNN) and graph neural network (GNN), towards predicting human ether-á-go-go related gene (hERG) derived toxicity. Using the largest hERG dataset derived to date, we have utilized 203,853 and 87,366 compounds for training and testing the models, respectively. The results show that GNN, SVM, XGBoost, DNN, RF, and GRU-DNN all performed well, with validation set AUC ROC scores equals 0.96, 0.95, 0.95, 0.94, 0.94 and 0.94, respectively. The GNN was found to be the top performing model based on predictive power and generalizability. The GNN technique is free of any feature engineering steps while having a minimal human intervention. The GNN approach may serve as a basis for comprehensive automation in predictive toxicology. We believe that the models presented here may serve as a promising tool, both for academic institutes as well as pharmaceutical industries, in predicting hERG-liability in new molecular structures. 

Place, publisher, year, edition, pages
Elsevier B.V. , 2023. Vol. 5, article id 100121
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:ri:diva-67450DOI: 10.1016/j.crtox.2023.100121Scopus ID: 2-s2.0-85171525959OAI: oai:DiVA.org:ri-67450DiVA, id: diva2:1803051
Note

This work has been funded by Research Institutes of Sweden RISE’s internal project “AI-TOX” (grant no. KFT SK-2021). Partial funding has been received from the “Safe and Efficient Chemistry by Design (SafeChem)” project funded by the Swedish Foundation for Strategic Environmental Research, MISTRA (grant no. DIA 2018/11 ).

Available from: 2023-10-06 Created: 2023-10-06 Last updated: 2024-02-19Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Chavan, SwapnilCotgreave, Ian

Search in DiVA

By author/editor
Chavan, SwapnilCotgreave, Ian
By organisation
Data ScienceChemical and Pharmaceutical Toxicology
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 11 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf