Accelerated Transfer Learning for Protein-Protein Interaction Prediction
Bibliographic record
Abstract
This thesis explores issues arising when one attempts to predict protein-protein interactions (PPI) involving multiple species using the Protein-protein Interaction Prediction Engine (PIPE) method.In cross-species predictions, where one predicts PPI in a target species given known PPI in a different training species, we showed that prediction performance is inversely correlated to the evolutionary distance between training and target 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.In inter-species predictions, one attempts to predict interactions between proteins arising from two different species, such as a host and a pathogen.For the first time, we have shown that PIPE is able to predict such inter-species PPI by predicting 229 novel PPI between HIV and human at an estimated precision of 82% (100:1 class imbalance).Lastly, by modifying a main data structure of PIPE, we also improved the speed of the PIPE algorithm by a factor of 53x when predicting H. sapiens PPI.Using the methods developed in this thesis, we have predicted all possible PPI between soybean and the Soybean Cyst Nematode pathogen.Collaborators at Agriculture and Agri-Food Canada will be pursuing and validating these predictions as they seek to combat this costly pest.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".