Loading publication...

Noodle always counts identifier-free feature events for service operation. With your permission, Google Analytics can also estimate audience and engagement. Search text, publication identifiers, cookies, IP addresses and browser fingerprints are never included in Noodle's first-party event counters.

Privacy details

Article

Neural networks to learn protein sequence-function relationships from deep mutational scanning data.

Proceedings of the National Academy of Sciences of the United States of America - 30 Nov 2021

Gelman Sam, Fahlberg Sarah A, Heinzelman Pete, Romero Philip A, Gitter Anthony

Abstract excerpt

The mapping from protein sequence to function is highly complex, making it challenging to predict how sequence changes will affect a protein's behavior and properties. We present a supervised deep learning framework to learn the sequence-function mapping from deep mutational scanning data and make predictions for new, uncharacterized sequence variants. We test multiple neural network architectures, including a...

Topics

Join the communities discussing this publication.

Algorithms
Biochemical Phenomena
Deep Learning
Machine Learning
Neural Networks, Computer

Methodology Data sources Editorial policy Privacy About Contact Helena Bioinformatics

Sequence Analysis, Protein

Structure-Activity Relationship

Source

Open in PubMed Open full text

Identifiers

PMID: 34815338
DOI: 10.1073/pnas.2104878118
PMCID: PMC8640744

Provenance

PubMed metadata and Literature Corpus relationships. Corpus snapshot updated 1 Sept 2026.

Community interpretation

Discuss the evidence

Ask a question, add context or challenge an interpretation. Community activity is separate from scientific evidence strength.

0

discussions

0

replies

Citations encouraged

Claims remain attributable.

No discussion yet

Be the first to add a careful, cited interpretation.

Related research

Semantic proximity does not establish scientific evidence.

Research trail→

Click a neighbor to travelStep 1 · 12 closest

Neural networks to learn protein sequence-function relationships from deep mutational scanning data.PMID 34815338Read ↗

Overcoming Extrapolation Challenges of Deep Learning by Incorporating Physics in Protein Sequence-Function ModelingSemantic proximity Read ↗

Overcoming extrapolation challenges of deep learning by incorporating physics in protein sequence-function modeling.Semantic proximity Read ↗

Protein multi-level structure feature-integrated deep learning method for mutational effect prediction.Semantic proximity Read ↗

MuLAN: Mutation-driven Light Attention Networks for investigating protein-protein interactions from sequencesSemantic proximity Read ↗

Graph masked self-distillation learning for prediction of mutation impact on protein-protein interactions.Semantic proximity Read ↗

3D-ΔΔG: A Dual-Channel Prediction Model for Protein-Protein Binding Affinity Changes Following Mutation Based on Protein 3D Structures.Semantic proximity Read ↗

Flattening the curve-How to get better results with small deep-mutational-scanning datasets.Semantic proximity Read ↗

Sequence-aware Prediction of Point Mutation-induced Effects on Protein-Protein Binding Affinity using Deep LearningSemantic proximity Read ↗

DGCddG: Deep Graph Convolution for Predicting Protein-Protein Binding Affinity Changes Upon Mutations.Semantic proximity Read ↗

DeepPNI: a language- and graph-based model for mutation-driven protein-nucleic acid binding energetics.Semantic proximity Read ↗

A protein dynamics-based deep learning model enhances predictions of fitness and epistasis.Semantic proximity Read ↗

MutPPI+: a multimodal framework for predicting mutation effects on protein-protein interactions via mutation-path-based data augmentation.Semantic proximity Read ↗

⌁Select a neighboring publication to make it the new centre.