To complement experimental efforts, machine learning-based computational methods are playing an important role to predict human-virus protein-protein interactions (PPIs). To predict interactions between human and viral proteins, we combine evolutionary sequence profile features of interacting proteins with a Siamese convolutional neural network (CNN) architecture and a multi-layer perceptron, allowing us to outperform various feature encodings-based machine learning and state-of-the-art prediction methods. As our main contribution, we introduce a transfer learning method that reliably predicts interactions in a target human-virus domain based on training in a source human-virus domain. Finally, we utilize our transfer learning method to predict human-SARS-CoV-2 PPIs, indicating that our predictions are topologically and functionally similar to experimentally known interactions.

- Journal
- Articles
- Author Center
- Society
Editorial Board
The ESMED Editorial Board is comprised of experts from around the world.
Meet the community
See what our members have been working on.
Membership
Join ESMED for access to member-only content, congress discounts, and more.
- Policy