IndependentStudy

Protein - Protein Interactions (PPI) cause vital processes such as growing, division or maintenance of a cell. It is an important topic in order to understand the functioning of any cell with itself and others. Even though the most reliable techniques are experimental for investigating the properties of different structures, numerical methods are much faster with negligible error. Nevertheless, predictions of PPI needs improvements in order to derive more accurate results faster. In this project, we are testing the capability of deep learning algorithms and their performances over protein protein interaction predictions. By using different optimization algorithms and Convolutional Neural Networks, we are tackling the PPI prediction problem as a classification problem and measure performance. We are aiming to compare the performances of different optimization algorithms, which can be considered as a challenge for sparse data. In addition, we are implementing a heuristic algorithm for structural matching in a fast manner. Main motivation is to calculate similarity scores in order to determine relevant protein interfaces for a given arbitrary protein. Doing that in a fast way provides docking methods to work far faster by negligible error.