2026
Selected: Explainable machine learning and deep learning models for predicting TAS2R-bitter molecule interactions
Journal of Molecular Graphics and Modelling142, 109187
The structure and dynamics of bitter taste receptors such as TAS2R46, and computational tools that predict which molecules they recognise.
Bitter taste receptors (TAS2Rs) are proteins that detect bitter compounds, in the mouth and in many other tissues. They are hard to study experimentally, which makes them a good fit for computational methods.
I have studied the internal mechanics of the bitter receptor TAS2R46 with molecular dynamics and network analysis, and I work with colleagues on tools that predict which molecules a receptor will recognise, from structure-based screening of protein–ligand complexes to explainable machine-learning models. Much of this work grew out of the EU project VIRTUOUS.
The human bitter taste receptor TAS2R46 holding strychnine. I study how this receptor moves, and help build models that predict which molecules bitter receptors recognise.
Marked: every residue with a heavy atom within 4.0 Å of strychnine, with the shortest distance, measured from the deposited coordinates.
Structure: PDB 7XP6 (Xu et al., Science 2022)
| # | Residue | Atoms (receptor – ligand) | Distance |
|---|---|---|---|
| 1 | Asn65 | OD1 – CAC | 3.0 Å |
| 2 | Tyr85 | CD1 – OAJ | 3.1 Å |
| 3 | Glu265 | OE2 – NAY | 3.2 Å |
| 4 | Trp88 | CG – CAE | 3.3 Å |
| 5 | Thr180 | OG1 – OAO | 3.4 Å |
| 6 | Leu62 | CD1 – CAC | 3.5 Å |
2026
Journal of Molecular Graphics and Modelling142, 109187
2024
Frontiers in Molecular Biosciences11, 1473675
2024
Scientific Reports14(1), 6296