HPiP: an R/Bioconductor package for predicting host–pathogen protein–protein interactions from protein sequences using ensemble machine learning approach
Bibliographic record
Abstract
Abstract Motivation Despite arduous and time-consuming experimental efforts, protein–protein interactions (PPIs) for many pathogenic microbes with their human host are still unknown, limiting our understanding of the intricate interactions during infection and the identification of therapeutic targets. Since computational tools offer a promising alternative, we developed an R/Bioconductor package, HPiP (Host–Pathogen Interaction Prediction) software with a series of amino acid sequence property descriptors and an ensemble machine learning classifiers to predict the yet unmapped interactions between pathogen and host proteins. Results Using severe acute respiratory syndrome coronavirus 1 (SARS-CoV-1) or the novel SARS-CoV-2 coronavirus-human PPI training sets as a case study, we show that HPiP achieves a good performance with PPI predictions between SARS-CoV-2 and human proteins, which we confirmed experimentally in human monocyte THP-1 cells, and with several quality control metrics. HPiP also exhibited strong performance in accurately predicting the previously reported PPIs when tested against the sequences of pathogenic bacteria, Mycobacterium tuberculosis and human proteins. Collectively, our fully documented HPiP software will hasten the exploration of PPIs for a systems-level understanding of many understudied pathogens and uncover molecular targets for repurposing existing drugs. Availability and implementation HPiP is released as an open-source code under the MIT license that is freely available on GitHub (https://github.com/BabuLab-UofR/HPiP) as well as on Bioconductor (http://bioconductor.org/packages/devel/bioc/html/HPiP.html). Supplementary information Supplementary data are available at Bioinformatics Advances online.
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 | 0.007 | 0.019 |
| Meta-epidemiology (narrow) | 0.006 | 0.003 |
| Meta-epidemiology (broad) | 0.005 | 0.003 |
| Bibliometrics | 0.004 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.008 | 0.005 |
| Research integrity | 0.002 | 0.006 |
| Insufficient payload (model declined to judge) | 0.068 | 0.073 |
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".