Publication Date
Fall 2024
Degree Type
Master's Project
Degree Name
Master of Science in Computer Science (MSCS)
Department
Computer Science
First Advisor
William Andreopoulos
Second Advisor
Leonard P Wesley
Third Advisor
Wendy Lee
Keywords
ESMFold, ColabFold, Toxin-Antitoxin Systems, Fine-tuning, Mutated Sequence Analysis, Protein Structure Prediction, Data Augmentaion, Bacterial Genomics, Structural Bioinformatics
Abstract
Generative AI models have vast applications and one such critical application explored in this study is protein structure prediction. The 3D structures of proteins determine their function. Our study mainly focuses on using generative AI models such as ESMFold and ColabFold to predict and examine naturally occurring and mutated sequences. The workflow begins with collecting antimicrobial resistance (AMR) and toxin-antitoxin (TA) protein data. The sequences are applied over pretrained AI models to predict protein structures. Following this, models are fine-tuned with original and mutated target datasets. A comparison of models’ performances is done using metrics such as root mean square error, predicted template modeling and predicted local distance difference test scores. This work lays the foundation for future study in leveraging generative AI models for novel protein structures prediction and drug discovery.
Recommended Citation
Rao, Kruthi Shankar, "Comparison of protein structures predicted by genAI tools in a zero-shot manner" (2024). Master's Projects. 1435.
https://scholarworks.sjsu.edu/etd_projects/1435
