arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.20573cs.LG
收藏

MolGraphBench:用于分子回归任务的GNN架构基准测试

MolGraphBench: A Benchmark of GNN Architectures for Molecular Regression Tasks

Rajan 1;Ishaan Gupta 2

Rajan, Ishaan Gupta

AI总结:

提出MolGraphBench基准,比较四种GNN模型在分子回归任务上的性能,发现GCN和GIN为最优架构,并指出GNN层类型应作为可调超参数。

摘要:

分子通常表示为SMILES字符串,可以轻松转换为手工设计的描述符或指纹(FP)用于分子性质预测。研究表明,SMILES可以转换为分子图 $G = (V, E)$,其中原子为节点 $(V)$,键为边 $(E)$。这些分子图随后可用于训练图神经网络(GNN)模型。尽管近年来GNN(现有和新架构)在分子性质预测中的应用激增,但仍缺乏严格的基准测试。我们提出了MolGraphBench,一个包含四种常用GNN模型的全面基准测试,用于分子性质预测。基准测试结果表明,基于绝对性能、训练效率、迁移学习和预测质量,图卷积网络(GCN)和图同构网络(GIN)是分子图回归任务的最优GNN架构。研究还表明,在融合(GNN-FP)框架中,分子指纹具有非互补性。此外,我们的GNN模型在三个数据集上取得了优于或与当前最先进GNN基线相当的性能(B3DB上GCN的RMSE为0.518,FreeSolv上GIN-FP的RMSE为1.022,RT数据集上GIN的MAE为63.783)。本研究的发现表明,GNN层类型应被视为可调超参数,而非固定设计选择,以实现更优性能。

英文摘要:

Molecules are often represented as SMILES strings, which can be readily converted to hand-crafted descriptors or fingerprints (FP) for molecular property prediction. Research has demonstrated that SMILES can be converted to molecular graphs $G = (V, E)$, with atoms as nodes $(V)$ and bonds as edges $(E)$. These molecular graphs can subsequently be used to train graph neural networks (GNN) models. Despite the recent surge in application of GNN (existing and novel architectures) for molecular property prediction, a rigorous benchmark is still lacking. We propose MolGraphBench, a comprehensive benchmark of four commonly used GNN models for molecular property prediction. Benchmarking results demonstrate graph convolutional network (GCN) and graph isomorphism networks (GIN) as the optimal GNN architectures for molecular graph regression tasks, based on absolute performance, training efficiency, transfer learning and prediction quality. The study also indicates the non-complementary nature of molecular fingerprints in the fusion (GNN-FP) framework. Furthermore, our GNN models achieved performance superior or comparable performance to current state-of-the-art GNN baselines across three datasets (GCN with RMSE of $0.518$ on B3DB, GIN-FP with RMSE of $1.022$ on FreeSolv and GIN with MAE of $63.783$ on RT datasets). Findings from this study indicate that type of GNN-layer, should be treated as a tunable hyperparameter rather than a fixed design choice to achieve superior performance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑