Reproduce “ATOM3D: Tasks On Molecules in Three Dimensions” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Raphael J.L. Townshend, Martin Vögele, Patricia Suriana, Alexander Derry, Alexander Powers, Yianni Laloudakis, Sidhika Balachandar, Bowen Jing and 5 more
ATOM3D provides eight tasks and tools for comparing learning from atom positions across small molecules, proteins, and RNA. It evaluates 3D CNNs, graph neural networks, and equivariant networks against baselines. On mutation-stability prediction, a 3D graph neural network reached AUROC 0.609 ± 0.011, versus 0.554 for a sequence-only baseline. Results vary by task and do not establish universal gains.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.