Title:	Predicting Bacterial Surface Proteins Using Machine Learning
DNr:	Berzelius-2025-14
Project Type:	LiU Berzelius
Principal Investigator:	Astrid von Mentzer <astrid.von.mentzer@gu.se>
Affiliation:	Göteborgs universitet
Duration:	2025-01-22 – 2025-08-01
Classification:	10610
Homepage:	https://vonmentzerlab.com/
Keywords:

Abstract

This project aims to use machine learning to identify potential surface proteins in bacterial genomes. Drawing on public genomic databases and annotation tools, we will gather and curate a training set of known surface proteins. We will then develop predictive models that consider protein sequence features, structural properties, and functional annotations. By applying these models to large-scale datasets, we aim to find proteins that might play key roles in bacterial interaction with their surroundings. This work aligns with the broader goals of the DDLS research initiative: advancing computational methods in the life sciences to gain insights into microbial function and support applications such as vaccine target identification and diagnostic development.