Multitask Learning for Six-Pack Toxicity Prediction
Journal
Annals of Computer Science and Information Systems
Part Of
Annals of Computer Science and Information Systems
Position Papers of the 20th Conference on Computer Science and Intelligence Systems
Journal Volume
44
Start Page
93
End Page
97
ISSN
2300-5963
ISBN
[9788397329188]
Date Issued
2025-10-15
Author(s)
Abstract
The assessment of the six-pack toxicity, the crucial six systems and organ toxicities, is vital for ensuring the safe use of chemicals. Computational models capable of providing reliable predictions are acceptable for regulatory use to replace animal testing. However, data scarcity issues hindered the development of prediction models. This study proposed the first application of multitask learning to the six-pack toxicity for addressing data scarcity issues. Five algorithms were implemented and compared. Results showed that the distinct chemical space of tasks impedes the learning of shared representation of conventional algorithms, with performance worse than baseline models. In contrast, the MTForestNet algorithm built on a biological readacross concept performed best, with 3.1% and 3.3% improvement on AUC and accuracy, respectively. These findings demonstrate that biologically informed multitask learning can effectively overcome data scarcity and enhance toxicity prediction.
Event(s)
20th Conference on Computer Science and Intelligence Systems, FedCSIS 2025
Subjects
biological readacross
distinct chemical space
MTForestNet
multitask learning
six-pack toxicity
Publisher
Polish Information Processing Society
Description
20th Conference on Computer Science and Intelligence Systems, FedCSIS 2025
14 September 2025 - 17 September 2025, in Kraków
14 September 2025 - 17 September 2025, in Kraków
Type
conference paper