arrow
Return

Explainable Machine Learning and Deep Learning Models for Predicting TAS2R-Bitter Molecule Interactions

delete2025-10-08
delete0
delete
OA
AI
F
Francesco Ferri
M
Marco Cannariato
L
Lorenzo Pallante
E
Eric A. Zizzi
M
Marcello Miceli
M
Marco A. Deriu *
DOI:10.1016/j.jmgm.2025.109187delete
deleteOriginal
deleteShare
deleteSave
View PDF
Abstract

Abstract

En 中文
• Two ML models (TML & GCN) predict bitter molecule-receptor interactions • TML and GCN achieve comparable performance across all evaluation metrics • Models identify molecular features driving specific TAS2R targeting • SHAP analysis reveals feature importance for TML model predictions • GNNExplainer and Grad-CAM provide visual explanations on molecular structure
AI Summary

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.

Journal

Journal of Molecular Graphics and Modelling cover
Journal of Molecular Graphics and Modelling
IF:
3
Papers:
521
Citations:
1.3W

Organization

D
department of mechanical and aerospace engineering
Scholars:
332
Papers: 160
Citations: 2