Explainable Machine Learning and Deep Learning Models for Predicting TAS2R-Bitter Molecule Interactions - Research focus on ligand-TAS2R interaction prediction

Aug 4, 2026

Upload by Aurelia

This work develops explainable machine-learning and deep-learning models to predict interactions between bitter molecules and TAS2Rs using experimentally validated data. Bitter taste is mediated by TAS2R bitter taste receptors, a GPCR family, and TAS2Rs also function in extra-oral tissues with disease relevance. Because in-vitro screening of TAS2R ligands is costly and time-consuming, ML and DL support experimental ligand/target selection and interpretation. Combined models aim to provide high performance, improved explainability, and guidance for designing novel bitterants tailored to specific TAS2Rs.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Introduction

Taste perception and TAS2R biology

Rationale for computational prediction

Limitations of experimental assays and available databases

POLITECNICO DI TORINO Repository ISTITUZIONALE Explainable Machine Learning and Deep Learning Models for Predicting TAS2R-Bitter Molecule Interactions Original Explainable Machine Learning and Deep Learning Models for Predicting TAS2R-Bitter Molecule Interactions / Ferri, Francesco; Cannariato, Marco; Pallante, Lorenzo; Zizzi, Eric A. ; Miceli, Marcello; Deriu, Marco A.. -In: JOURNAL OF MOLECULAR GRAPHICS & MODELLING. -ISSN 1093-3263. -142:(2026) . [10 . 1016/j.jmgm.2025. 109187] Availability: This version is available at: 11583/3004052 since: 2025-10-15T10:45:45Z Publisher: Elsevier Published DOI:10.1016/j.jmgm.2025.109187 Terms of use: This article is made available under terms and conditions as specified in the corresponding bibliographic description in the repository Publisher copyright (Article begins on next page) 21 February 2026 Journal of Molecular Graphics and Modelling 142 (2026) 109187 Contents lists available at ScienceDirect Journal of Molecular Graphics and Modelling journal [homepage: www.elsevier.com/locate/jmgm](homepage: www.elsevier.com/locate/jmgm) | Explainable machine learning and deep learning models for predicting TAS2R-bitter molecule interactions
Francesco Ferria,b,1, Marco Cannariatoa,1, Lorenzo Pallantea, Eric A. Zizzia, Marcello Micelia, Marco A. Deriua,*
a Politecnico di Torino, PolitoBIOMedLab, Department of Mechanical and Aerospace Engineering, Torino, 10129, Italy
b TUM School of Life Sciences Weihenstephan, Technical University of Munich, Alte Akademie 8, 85354, Freising, Germany | | --- | | A B S T R A C T | | This work aims to develop explainable models to predict the interactions between bitter molecules and TAS2Rs via traditional machine-learning and deep-learning methods starting from experimentally validated data. Bitterness is one of the five basic taste modalities that can be perceived by humans and other mammals. It is mediated by a family of G protein-coupled receptors (GPCRs), namely taste receptor type 2 (TAS2R) or bitter taste receptors. Furthermore, TAS2Rs participate in numerous functions beyond the gustatory system and have implications for various diseases due to their expression in various extra-oral tissues. For this reason, predicting the specific ligand-TAS2Rs interactions can be useful not only in the field of taste perception but also in the broader context of drug design. Considering that in-vitro screening of potential TAS2R ligands is expensive and time-consuming, machine learning (ML) and deep learning (DL) emerged as powerful tools to assist in the selection of ligands and targets for experimental studies and enhance our understanding of bitter receptor roles. In this context, ML and DL models developed in this work are both characterized by high performance and easy applicability. Furthermore, they can be synergistically integrated to enhance model explainability and facilitate the interpretation of results. Hence, the presented models promote a comprehensive understanding of the molecular characteristics of bitter compounds and the design of novel bitterants tailored to target specific TAS2Rs of interest. | 1. Introduction Taste perception is a crucial determinant of food intake and consumption patterns [1], with substantial consequences for human nutrition and health [2]. Several pathological conditions can impair the sense of taste and lead to alterations in quality of life and body weight regulation [3]. The molecular mechanisms of bitter taste perception involve a subfamily of 25 G protein-coupled receptors (GPCRs), called type 2 taste receptors (TAS2Rs) [4]. These proteins are also expressed in various extraoral tissues, where they may perform additional physiological functions, such as modulating inflammatory response, controlling upper respiratory immunity and others [5,6]. TAS2Rs exhibit a remarkable diversity and specificity in their ligand recognition and activation. Some TAS2Rs are defined as promiscuou

FAQs

What is the main goal of the proposed work?

To develop explainable machine-learning and deep-learning models that predict interactions between bitter molecules and TAS2Rs using experimentally validated data.

Why are TAS2R-ligand interactions important beyond taste perception?

TAS2Rs are expressed in extra-oral tissues and are implicated in functions and diseases, making interaction prediction valuable for broader biomedical contexts including drug design.

How do the authors address the cost of in-vitro TAS2R ligand screening?

They use ML and DL models to assist ligand and target selection for experimental studies, improving efficiency and supporting interpretation of results.

Aug 4, 2026

Upload by Aurelia