Project Abstract

Motivated by the need to unravel the intricate roles of proteins for biomedical improvements, this study explores the use of machine learning to reliably predict protein activities. Given the critical role that proteins play in biological processes and disease mechanisms, there has been an increase in demand for efficient computational methodologies, particularly in light of growing protein data and the constraints of traditional experimental methods. This study uses machine learning and bioinformatics tools to revolutionise protein function prediction by combining structural, sequence, and evolutionary data, ultimately aiding drug discovery and medicinal discoveries. The unique proposition is the combination of multi-omics data to improve predictive models, which promises a full understanding of protein activities across many biological situations. The study uses a novel combination of SVMs, random forests, and deep learning architectures to construct robust prediction models capable of exact protein function annotation, paving the path for expedited scientific research and personalized therapy. The methodology involves data gathering from protein databases, preprocessing, feature extraction, model training, and evaluation, with an emphasis on scalability, efficiency, and generalization. The major findings are expected to show the usefulness of machine learning models in reliably predicting protein functions using multi-omics data, resulting in innovative insights and faster drug discovery processes. In conclusion, this study is set to improve the scientific community’s grasp of protein biology and provide innovative solutions to important biomedical concerns, eventually benefiting human health and well-being.

Keywords: Machine Learning, Bioinformatics, Python Algorithms

 Conference Details

Session: Presentation Stream 31 at Presentation Slot 8

Location: GH018 at Wednesday 8th 13:30 – 17:00

Markers: Gregory Cheng, Ehinafa Akinola (GTA)

Course: MSc Advanced Computer Science, Masters PG

Future Plans: I’m looking for an industry placement