AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

End-to-End Learning on 3D Protein Structure for Interface Prediction

NeurIPS 2019 · Probabilistic and scientific modeling › Molecules and proteins · approach: Protein interface and docking prediction · 41 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
Use end-to-end SASNet trained on DIPS rather than the strongest hand-engineered baseline BIPSPI trained on DB5Paired protein interface prediction · DB5-testCAUROC: 0.878 -> 0.892Helped
Train Node Average on DIPS instead of DB5-trainNode Average · DB5-testCAUROC: 0.876 -> 0.712Hurt
Train BIPSPI on DIPS instead of DB5-trainBIPSPI · DB5-testCAUROC: 0.878 -> 0.836Hurt
Train SASNet on DIPS instead of DB5-trainSASNet · DB5-testCAUROC: 0.876 -> 0.892Helped
Increase the voxel grid edge size, keeping resolution at 1 Å (ablation)SASNet · DB5-testCAUROCHelped
Increase the DIPS training dataset size (ablation)SASNet · DB5-testCAUROCHelped
Increase grid edge size beyond 27 Å (ablation)SASNet · DB5-testCAUROCNo clear effect
Filter DIPS training examples with 20% or greater sequence identity to DB5 (ablation)SASNet · DB5-testCAUROC: 0.887No clear effect
Remove DIPS complexes sharing domain-domain interactions with DB5 (ablation)SASNet · DB5-testCAUROC: 0.883No clear effect

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