CurvFlow-DTA: dual-graph discrete Ricci curvature flow for drug--target affinity prediction
Abstract
Graph neural networks are widely used for drug--target affinity (DTA) prediction, and discrete Ricci curvature has recently been used to characterize molecular graph geometry. Existing curvature-aware DTA approaches mainly use static curvature on the drug graph while representing proteins primarily with sequence-derived features. This leaves pair-adaptive use of graph geometry underexplored, which may limit adaptation to unseen entities in cold-start settings relevant to practical screening. We present CurvFlow-DTA, which replaces a single static curvature representation with weighted Forman curvature flow on both molecular and protein residue--residue contact graphs. A label-independent flow trajectory is precomputed for each entity, and a pair-conditioned selector determines the horizons read by a dual-branch Flow-GINE. A frozen ESM-2 supplies residue-level representations and contact scores used to construct the protein graph. Inference requires only SMILES strings and protein sequences, without a bound complex structure. On Davis and KIBA, CurvFlow-DTA improves on the protocol-matched Ricci-GraphDTA baseline in every warm and cold-start setting. Warm-split mean squared error (MSE) decreases by 19.9% on Davis and 18.9% on KIBA. Across the six cold-start comparisons, MSE decreases by $14.3−−27.4\%$, with higher concordance index (CI) throughout. Within our compiled set of literature baselines, CurvFlow-DTA achieves the lowest MSE on both warm benchmarks and across four out of six cold-start evaluation settings.
Related
- Chemical and geometric representation fidelity improves drug--target affinity prediction
- A semi-discrete scheme for horizontal mean curvature flow in the Heisenberg group
- Discrete Isoperimetric Inequalities via Curvature
- Curvature-Dimension Conditions Along Discrete Refinement Sequences
- Structural limit of volatility target under stochastic volatility and discrete corrections
BibTeX
@misc{ma2026curvflowdta,
title = {{CurvFlow-DTA: dual-graph discrete Ricci curvature flow for drug--target affinity prediction}},
author = {Jicheng Ma and Yunyan Yang and Juan Zhao and Liang Zhao},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.22862},
eprint = {2609.22862},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.22862},
}
RIS
TY - UNPB
TI - CurvFlow-DTA: dual-graph discrete Ricci curvature flow for drug--target affinity prediction
AU - Jicheng Ma
AU - Yunyan Yang
AU - Juan Zhao
AU - Liang Zhao
PY - 2026
DA - 2026/09/19
PB - arXiv
DO - 10.48550/arxiv.2609.22862
UR - https://arxiv.org/abs/2609.22862
AB - Graph neural networks are widely used for drug--target affinity (DTA) prediction, and discrete Ricci curvature has recently been used to characterize molecular graph geometry. Existing curvature-aware DTA approaches mainly use static curvature on the drug graph while representing proteins primarily with sequence-derived features. This leaves pair-adaptive use of graph geometry underexplored, which may limit adaptation to unseen entities in cold-start settings relevant to practical screening. We present CurvFlow-DTA, which replaces a single static curvature representation with weighted Forman curvature flow on both molecular and protein residue--residue contact graphs. A label-independent flow trajectory is precomputed for each entity, and a pair-conditioned selector determines the horizons read by a dual-branch Flow-GINE. A frozen ESM-2 supplies residue-level representations and contact scores used to construct the protein graph. Inference requires only SMILES strings and protein sequences, without a bound complex structure. On Davis and KIBA, CurvFlow-DTA improves on the protocol-matched Ricci-GraphDTA baseline in every warm and cold-start setting. Warm-split mean squared error (MSE) decreases by 19.9% on Davis and 18.9% on KIBA. Across the six cold-start comparisons, MSE decreases by $14.3−−27.4\%$, with higher concordance index (CI) throughout. Within our compiled set of literature baselines, CurvFlow-DTA achieves the lowest MSE on both warm benchmarks and across four out of six cold-start evaluation settings.
ER -