SIGNATURE WORK
CONFERENCE & EXHIBITION 2023

“Machine learning techniques have been used to greatly expand computational biology research in recent years. In this study, a multi-input convolutional neural network on graph-represented polymer data (OR-GRCNN) is used to predict the binding affinity of olfactory receptor proteins and chemicals. Analysis of the OR-GRCNN model’s performance demonstrates that it performs better than other binary classification models in the area. In order to encourage model reuse, OR-Python GRCNN’s repository follows clean coding norms and includes docstrings outlining code functionality in each primary module. In addition, a survey on readability and reuse of code in the field of computational biology is carried out. It shows that just around half of the top monthly publications in the IEEE/ACM Transactions on Computational Biology journal for the year 2022 provided their source code.

This project has two effects: The computational biology code survey shows that there is no association between a high citation count and open source code released in the IEEE/ACM Transactions on Computational Biology journal. OR-GRCNN predicts OR-ligand binding using a unique memory-efficient protein graph representation.”