Embedded Files
Improving network inference for protein structural classification
Improving network inference for protein structural classification
Student: Jennifer Morones
Studying proteins in biology is vital for the understanding of their functions, however characterizing them through wet experiments is time and resource consuming. Protein Structure Networks (PSN) are powerful models of 3D structures of proteins, as they focus on interaction and pathways between the amino acids. Protein structural classification (PSC) is a well-established computational task that involves categorizing proteins into predefined structural classes based on their sequence or 3D structural features. However, static PSNs using PSC approaches often overlook the dynamic of the protein folding process, limiting accuracy.
The overall goal of the project is to analyze further improvements on protein structural classification leveraging on the protein dynamics.Â
Dr Tijana Milenkovic tackles the flood of data about individuals, e.g., their molecular, social interaction, physical activity, behavioral, physiological, or electronic health record data. Prof. Milenkovic aims to develop novel computational approaches (algorithms) for modeling, integrating, and mining such data from a network perspective, and in interdisciplinary collaborations, to use the algorithms to give personalized feedback to individuals about improving their health.
Prof. Milenkovic is committed to equity, diversity, and inclusion activities intended to improve experience of gender, ethnic, and other minorities in the field of computing.
Page updated
Report abuse