• English
    • français
  • Help
  •  | 
  • Contact
  •  | 
  • About
  •  | 
  • Login
  • HAL portal
  •  | 
  • Pages Pro
  • EN
  •  / 
  • FR
View Item 
  •   LillOA Home
  • Liste des unités
  • Protéomique, Réponse Inflammatoire, Spectrométrie de Masse (PRISM) - U1192
  • View Item
  •   LillOA Home
  • Liste des unités
  • Protéomique, Réponse Inflammatoire, Spectrométrie de Masse (PRISM) - U1192
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Toxic colors: the use of deep learning for ...
  • BibTeX
  • CSV
  • Excel
  • RIS

Document type :
Article dans une revue scientifique: Article original
DOI :
10.1021/acs.jcim.8b00338
PMID :
30063345
Permalink :
http://hdl.handle.net/20.500.12210/74968
Title :
Toxic colors: the use of deep learning for predicting toxicity of compounds merely from their graphic images
Author(s) :
Fernandez, Michael [Auteur]
Ban, Fuqiang [Auteur]
Woo, Godwin [Auteur]
Hsing, Michael [Auteur]
Yamazaki, Takeshi [Auteur]
Leblanc, Eric [Auteur] refId
Rennie, Paul S. [Auteur]
Welch, William J. [Auteur]
Cherkasov, Artem [Auteur]
Journal title :
Journal of Chemical Information and Modeling
Abbreviated title :
J Chem Inf Model
Publication date :
2018-07-31
ISSN :
1549-960X
HAL domain(s) :
Sciences du Vivant [q-bio]
English abstract : [en]
The majority of computational methods for predicting toxicity of chemicals are typically based on "nonmechanistic" cheminformatics solutions, relying on an arsenal of QSAR descriptors, often vaguely associated with chemical ...
Show more >
The majority of computational methods for predicting toxicity of chemicals are typically based on "nonmechanistic" cheminformatics solutions, relying on an arsenal of QSAR descriptors, often vaguely associated with chemical structures, and typically employing "black-box" mathematical algorithms. Nonetheless, such machine learning models, while having lower generalization capacity and interpretability, typically achieve a very high accuracy in predicting various toxicity endpoints, as unambiguously reflected by the results of the recent Tox21 competition. In the current study, we capitalize on the power of modern AI to predict Tox21 benchmark data using merely simple 2D drawings of chemicals, without employing any chemical descriptors. In particular, we have processed rather trivial 2D sketches of molecules with a supervised 2D convolutional neural network (2DConvNet) and demonstrated that the modern image recognition technology results in prediction accuracies comparable to the state-of-the-art cheminformatics tools. Furthermore, the performance of the image-based 2DConvNet model was comparatively evaluated on an external set of compounds from the Prestwick chemical library and resulted in experimental identification of significant and previously unreported antiandrogen potentials for several well-established generic drugs.Show less >
Language :
Anglais
Audience :
Internationale
Popular science :
Non
Administrative institution(s) :
INSERM
Université de Lille
Collections :
  • Protéomique, Réponse Inflammatoire, Spectrométrie de Masse (PRISM) - U1192
Submission date :
2022-06-15T13:57:36Z
Université de Lille

Mentions légales
Université de Lille © 2017