AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Learning residue level protein dynamics with multiscale Gaussians

Mihir Bafna, Bowen Jing, Bonnie Berger

ICLR 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Generative protein structure design

Read the paperCheck your idea against 137,217 papers

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
replace log-Frobenius loss with MSE (ablation)DynaProt-M · ATLAS marginalsRMSF r / sym KL: RMSF r 0.87 -> 0.38; Sym KL 0.91 -> 9.26Hurt
replace SE(3)-invariant IPA blocks with MLPs (ablation)DynaProt-M · ATLAS marginalsRMSF r / RMWD: RMSF r 0.87 -> 0.48; RMWD 1.18 -> 1.92Hurt
explicit marginal prediction (DynaProt-M) vs FlexPert-3D with 1.2B paramsDynaProt-M · ATLAS RMSFRMSF r: 0.830 -> 0.865Improved
DynaProt-M vs NMADynaProt-M · ATLAS marginalsRMWD / Sym KL: 1.45 -> 1.18; 4.56 -> 0.91Improved
DynaProt-M vs AFMD+TemplatesDynaProt-M · ATLAS marginalsRMWD median: 1.10 (AFMD+T) vs 1.18No clear effect
DynaProt-J vs NMA on residue couplingDynaProt-J · ATLAS couplingpeak Pearson r: 0.59 -> 0.71Improved
DynaProt joint-reconstruction ensemble vs AFMD+TDynaProt · ATLAS ensemblesglobal RMSF r / MD PCA W2: global RMSF r 0.91 -> 0.71; MD PCA W2 1.25 -> 1.74Mixed
DynaProt vs NMA ensemblesDynaProt · ATLAS ensemblestransient contacts J: 0.33 -> 0.29Hurt

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