Accelerated Transfer Learning for Protein-Protein Interaction Prediction

Public Deposited
Resource Type
  • This thesis explores issues arising when predicting protein-protein interactions (PPI) involving multiple species with the Protein-protein Interaction Prediction Engine (PIPE). When predicting one species' PPI from another's, we showed that prediction performance is inversely correlated to the evolutionary distance between training and testing species. With a change in the score calculation, we improved the area under the precision-recall curve by 45% when using seven well-studied species to predict an eighth. We then showed that PIPE was able to predict PPI between species by predicting 229 novel PPI between HIV and human at an estimated precision of 82% (100:1 class imbalance). By modifying a main data structure, we also improved the speed of the PIPE algorithm by a factor of 53x when predicting H. sapiens PPI. Using these best practices, we predicted all possible PPI between soybean and its costly pest, the Soybean Cyst Nematode, for our collaborators at Agriculture Canada.

Thesis Degree Level
Thesis Degree Name
Thesis Degree Discipline
Rights Notes
  • Copyright © 2018 the author(s). Theses may be used for non-commercial research, educational, or related academic purposes only. Such uses include personal study, research, scholarship, and teaching. Theses may only be shared by linking to Carleton University Institutional Repository and no part may be used without proper attribution to the author. No part may be used for commercial purposes directly or indirectly via a for-profit platform; no adaptation or derivative works are permitted without consent from the copyright owner.
Date Created
  • 2018


In Collection: