AI-guided prioritization of potential FAM171A2 ligands through pharmacophore-aware and receptor-based screening
- 1Department of Anesthesia, The Second Affiliated Hospital of Xi'an Jiaotong University 710004 China
- 2Department of Bioengineering, College of Life Science, Dalian Minzu University 116600 China
Abstract
FAM171A2 has recently emerged as a potential therapeutic target in Parkinson's disease because it selectively recognizes α-synuclein fibrils and promotes their neuronal uptake. However, ligand prioritization for this receptor remains challenging because of scarce known binders, the absence of an experimentally resolved ligand-bound structure and the requirement for blood–brain barrier permeability. Here, we established an AI-guided computational screening framework that integrates pharmacophore-aware retrieval, hierarchical receptor-based screening, blood–brain barrier permeability prediction and molecular dynamics-based assessment to prioritize potential FAM171A2 ligands. Starting from 1.41 million compounds, the workflow progressively narrowed the search space to two structurally representative candidates, E634-1386 and F707-2364. Both candidates showed more favorable in silico profiles than the reference blocker Bemcentinib across complementary computational assessments: E634-1386 achieved higher KarmaDock and CarsiDock scores, whereas F707-2364 yielded the most favorable AutoDock Vina score and MM/GBSA estimate. Molecular dynamics simulations further suggested stable predicted FAM171A2-bound complexes for both candidates, with lower complex RMSD, reduced radius of gyration and lower protein solvent-accessible surface area relative to Bemcentinib, while maintaining broadly similar residue-level flexibility patterns. Representative interaction analysis suggested that E634-1386 may rely more on polar contacts, whereas F707-2364 may engage the predicted binding region through broader hydrophobic complementarity. Together, these results prioritize E634-1386 and F707-2364 as computationally selected, experimentally testable FAM171A2 ligand candidates and provide a practical strategy for early-stage ligand prioritization against structurally challenging central nervous system targets.
Related
- A Pharmacophore-Based Virtual Screening Approach for the Discovery of Novel P2Y ₁₂ Receptor Ligands
- Integrated machine learning, pharmacophore screening, and structure-based computational approaches for the identification of novel JAK3 inhibitors
- Integrated Computational Design and Multistage Virtual Screening for the Prioritization of Putative USP5 Catalytic-Cleft Inhibitors
- Computational design and prioritization of database-absent small-molecule GLP-1 receptor agonist candidates from public bioactivity data
- Chemistry-informed non-covalent graphs and pharmacophore supervision for molecular property prediction
BibTeX
@article{yan2026aiguided,
title = {{AI-guided prioritization of potential FAM171A2 ligands through pharmacophore-aware and receptor-based screening}},
author = {Baihui Yan and Yang Lu and Yanni Tong and Shaolong Lin and Wei Li},
journal = {RSC Advances},
year = {2026},
month = nov,
doi = {10.1039/d6ra07609b},
url = {https://doi.org/10.1039/d6ra07609b},
note = {CC BY-NC 3.0},
}
RIS
TY - JOUR
TI - AI-guided prioritization of potential FAM171A2 ligands through pharmacophore-aware and receptor-based screening
AU - Baihui Yan
AU - Yang Lu
AU - Yanni Tong
AU - Shaolong Lin
AU - Wei Li
PY - 2026
DA - 2026/11/12
JO - RSC Advances
DO - 10.1039/d6ra07609b
UR - https://doi.org/10.1039/d6ra07609b
AB - FAM171A2 has recently emerged as a potential therapeutic target in Parkinson's disease because it selectively recognizes α-synuclein fibrils and promotes their neuronal uptake. However, ligand prioritization for this receptor remains challenging because of scarce known binders, the absence of an experimentally resolved ligand-bound structure and the requirement for blood–brain barrier permeability. Here, we established an AI-guided computational screening framework that integrates pharmacophore-aware retrieval, hierarchical receptor-based screening, blood–brain barrier permeability prediction and molecular dynamics-based assessment to prioritize potential FAM171A2 ligands. Starting from 1.41 million compounds, the workflow progressively narrowed the search space to two structurally representative candidates, E634-1386 and F707-2364. Both candidates showed more favorable in silico profiles than the reference blocker Bemcentinib across complementary computational assessments: E634-1386 achieved higher KarmaDock and CarsiDock scores, whereas F707-2364 yielded the most favorable AutoDock Vina score and MM/GBSA estimate. Molecular dynamics simulations further suggested stable predicted FAM171A2-bound complexes for both candidates, with lower complex RMSD, reduced radius of gyration and lower protein solvent-accessible surface area relative to Bemcentinib, while maintaining broadly similar residue-level flexibility patterns. Representative interaction analysis suggested that E634-1386 may rely more on polar contacts, whereas F707-2364 may engage the predicted binding region through broader hydrophobic complementarity. Together, these results prioritize E634-1386 and F707-2364 as computationally selected, experimentally testable FAM171A2 ligand candidates and provide a practical strategy for early-stage ligand prioritization against structurally challenging central nervous system targets.
SN - 2046-2069
ER -