Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Correlated clustering and projection for dimensionality reduction.

Machine learning: science and technology·2026

Same author

VARIANT: Web Server for Decoding and Analyzing Viral Mutations at Genome and Protein Levels.

ArXiv·2026

Same author

Manifold topological deep learning for biomedical data.

Nature communications·2026

Same author

A review of recent advances in generative artificial intelligence models for biomolecular sciences.

Acta pharmaceutica Sinica. B·2026

Same author

CAP: Commutative algebra prediction of protein-nucleic acid binding affinities.

Machine learning: science and technology·2026

Same author

Topology-preserving Hodge decomposition in the Eulerian representation.

Beijing journal of pure & applied mathematics·2026

Same journal

Gaining biological insights through supervised data visualization.

Nature computational science·2026

Same journal

The inequalities of GPU access.

Nature computational science·2026

Same journal

Social technologies need societal alignment.

Nature computational science·2026

Same journal

The Quantum Optimization Benchmarking Library.

Nature computational science·2026

Same journal

Setting benchmarks for practical quantum utility of combinatorial optimization.

Nature computational science·2026

Same journal

Evidence of scaling advantage on an NP-complete problem with enhanced quantum solvers.

Nature computational science·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Jul 18, 2025

Engineering 'Golden' Fluorescence by Selective Pressure Incorporation of Non-canonical Amino Acids and Protein Analysis by Mass Spectrometry and Fluorescence

Engineering 'Golden' Fluorescence by Selective Pressure Incorporation of Non-canonical Amino Acids and Protein Analysis by Mass Spectrometry and Fluorescence

Published on: April 27, 2018

Persistent spectral theory-guided protein engineering.

Yuchi Qiu¹, Guo-Wei Wei^1,2,3

¹Department of Mathematics, Michigan State University, East Lansing, MI 48824, USA.

Nature Computational Science

|August 28, 2023

Summary

This summary is machine-generated.

This study introduces TopFit, a novel framework using topology to enhance protein engineering by analyzing protein structure and sequence. TopFit captures shape evolution and sequence differences to predict protein fitness more accurately.

More Related Videos

Residue-Specific Exchange of Proline by Proline Analogs in Fluorescent Proteins: How "Molecular Surgery" of the Backbone Affects Folding and Stability

Residue-Specific Exchange of Proline by Proline Analogs in Fluorescent Proteins: How "Molecular Surgery" of the Backbone Affects Folding and Stability

Published on: February 3, 2022

Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins

Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins

Published on: July 8, 2025

Related Experiment Videos

Last Updated: Jul 18, 2025

Engineering 'Golden' Fluorescence by Selective Pressure Incorporation of Non-canonical Amino Acids and Protein Analysis by Mass Spectrometry and Fluorescence

Engineering 'Golden' Fluorescence by Selective Pressure Incorporation of Non-canonical Amino Acids and Protein Analysis by Mass Spectrometry and Fluorescence

Published on: April 27, 2018

Residue-Specific Exchange of Proline by Proline Analogs in Fluorescent Proteins: How "Molecular Surgery" of the Backbone Affects Folding and Stability

Residue-Specific Exchange of Proline by Proline Analogs in Fluorescent Proteins: How "Molecular Surgery" of the Backbone Affects Folding and Stability

Published on: February 3, 2022

Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins

Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins

Published on: July 8, 2025

Area of Science:

Computational Biology
Protein Engineering
Structural Bioinformatics
Machine Learning

Background:

Protein engineering optimizes protein function but is limited by experimental screening capacity.
Machine learning accelerates protein engineering, yet 3D structural complexity impedes deep mutational screening.
Existing topological methods like persistent homology struggle to capture dynamic shape changes in protein data.

Purpose of the Study:

To develop a novel framework, Topology-offered protein Fitness (TopFit), to integrate protein sequence and structure information for improved fitness prediction.
To address limitations of current methods in capturing mutation-induced topological and shape variations in proteins.
To enhance the accuracy of predicting protein fitness landscapes through advanced topological and machine learning approaches.

Main Methods:

Developed the TopFit framework, integrating persistent spectral theory and a novel topological Laplacian.
Employed an ensemble regression strategy combining topological invariants, shape evolution analysis, and sequence embeddings.
Utilized two auxiliary sequence embeddings to capture sequence disparity alongside structural topological features.

Main Results:

TopFit successfully captures mutation-induced topological invariants and shape evolution within protein fitness landscapes.
The framework integrates sequence and structural topological information for comprehensive protein fitness prediction.
Evaluated on 34 benchmark datasets with 128,634 variants, demonstrating robust performance across diverse protein structures and dataset sizes.

Conclusions:

The TopFit framework offers a significant advancement in protein engineering by leveraging topological data analysis.
TopFit complements existing sequence and structure-based embeddings, providing a more holistic view of protein fitness.
This approach holds promise for accelerating the design and optimization of proteins with desired functionalities.