AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

ProteinNPT: Improving Protein Property Prediction and Design with Non-Parametric Transformers

Pascal Notin, Ruben Weitzman, Debora Susan Marks, Yarin Gal

NeurIPS 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Protein fitness and variant effect prediction · 24 citations (OpenAlex)

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
add column-wise attention across labeled sequences (ablation)ProteinNPT · 8 DMS assays, Contiguous CVSpearman: 0.425 -> 0.501Improved
add column-wise attention across labeled sequences (ablation)ProteinNPT · 8 DMS assays, Random CVSpearman: 0.669 -> 0.684Improved
add column-wise attention across labeled sequences (ablation)ProteinNPT · 8 DMS assays, Modulo CVSpearman: 0.530 -> 0.531No clear effect
use labeled sequences at inference (100 vs 0) (ablation)ProteinNPT · 8 DMS assays, AverageSpearman: 0.317 -> 0.569Improved
increase labeled sequences at inference beyond 1000 (ablation)ProteinNPT · 8 DMS assays, AverageSpearman: 0.572 -> 0.572No clear effect
replace one-hot features with MSA Transformer embeddings (augmented)ridge/baseline regressors · ProteinGym Contiguous CVSpearman: 0.41 -> 0.47Improved
augment one-hot ridge regression with unsupervised fitness predictionsOHE baseline · ProteinGym Contiguous CVSpearman: 0.08 -> 0.40Improved
replace Embed.-Aug. baseline with ProteinNPTEmbed. - Aug. (MSAT) · ProteinGym Random CVSpearman: 0.57 -> 0.65Improved
replace Embed.-Aug. baseline with ProteinNPTEmbed. - Aug. (MSAT) · ProteinGym Contiguous CVSpearman: 0.47 -> 0.48No clear effect
one-hot encodings used for cross-position extrapolationOHE · ProteinGym Contiguous CVSpearman: 0.54 (Random) -> 0.08 (Contiguous)Hurt
add auxiliary labels (unsupervised fitness predictions) (ablation)ProteinNPT · ProteinGymSpearmanImproved

Later papers that compared with this method

4 later accepted papers report ProteinNPT as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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