AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

PLA-MGRA: Multi-Granularity and Relation-Aware Learning for Efficient and Generalizable Protein-Ligand Binding Affinity Prediction

Shunfan Li, Jiangkai Long, Xin Zou, Chang Tang, Yuanyuan Liu, Xiao He, Xuesong Yan

AAAI 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Protein interface and docking prediction · 0 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add multi-granularity learning (top-k coarse abstraction plus fine-grained restoration) (ablation)PLA-MGRA · PDBbind v2013/v2016/v2019 test setsRMSE and Pearson correlationImproved
add relation-aware learning (mutual ligand/protein-induced attention) (ablation)PLA-MGRA · PDBbind v2019 holdout setRMSE and Pearson correlationImproved
add both multi-granularity and relation-aware modules to the GIGN encoder (ablation)GIGN encoder · PDBbind v2013/v2016/v2019 test setsRMSE and Pearson correlationImproved
use GIGN encoder instead of GCN encoder (ablation)PLA-MGRA · PDBbind v2013/v2016/v2019 test setsRMSE and Pearson correlationImproved
use PLA-MGRA instead of SS-GNN on temporal-split 2019 holdoutprotein-ligand affinity models · PDBbind 2019 holdout setRMSE: 1.450 -> 1.357Improved
use small interaction-based model (1.83M parameters) instead of large pretrained EGNN-PLM (650M)PLA 30% split · PLA 30%RMSE: 1.403 -> 1.387Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy