Machine learning meets and enhances protein engineering

Machine learning plays a pivotal role in modern science, and the study of proteins is equally vital due to their essential biological functions. In my research, I propose new machine learning models and algorithms, including novel generative models, advanced data augmentation techniques, innovative model architecture designs, and optimized loss functions. These innovations are meticulously applied to the field of protein research, aiming to enhance the accuracy and efficiency of protein analysis and prediction. The proposed methodologies offer significant improvements over traditional approaches, demonstrating the transformative potential of integrating machine learning with protein science. These techniques have been successfully applied to various protein-related tasks, significantly enhancing the models' ability to generalize from limited data and other specific training settings. These tasks encompass predicting the effects of mutations, forecasting enzyme functions, and estimating binding affinities, among others. Through my work, I have significantly enhanced the performance of previous methodologies, establishing new state-of-the-art benchmarks across these areas.

Name thesis:degree_name

Doctor of Philosophy

Discipline thesis:degree_discipline

Computer Science

Grantor

The University of Texas at Austin

Year dc:date.issued

2024

Author dc:creator

• Gong, Chengyue

Advisor dc:contributor.advisor

• Liu, Qiang (Ph. D. in computer science)

Committee members dc:contributor.committeemember

• Adam Klivans
• Mingyuan Zhou
• Qixing Huang

• Machine learning
• Generative model
• Protein engineering

Language dc:language.iso

English

Identifier URI

https://doi.org/10.26153/tsw/56640

OAI identifier oai:identifier

oai:repositories.lib.utexas.edu:2152/130141