Congratulations to Palina on her new paper in the Journal of Chemical Information and Modeling journal (https://doi.org/10.1021/acs.jcim.5c00336). The study looks at how well current computational tools can predict ligand binding sites in the protein–membrane interface region. Using datasets of GPCRs and ion channels, the paper benchmarks methods, from geometry-based tools like Fpocket to deep learning approaches like DeepPocket and PUResNetV2.0. Although some methods performed reasonably well, all of them struggled compared to results on soluble proteins, highlighting the need for better tools tailored to membrane-embedded binding sites.
New paper in the JCIM published by Palina
