Applying integrative geneset-embedded non-negative matrix factorization to discovery of biomarkers for major depressive disorder antidepressant response
Bibliographic record
Abstract
Non-negative matrix factorization (NMF) is a dimension reduction technique that is capable of deriving complex latent themes from multi-modal inputs while retaining interpretability due to its non-negative nature. The technique has been applied to various fields, and we have developed a NMF-based tool for multi-omic integration and applied the model to a major depressive disorder (MDD) patient cohort to investigate important players in antidepressant (AD) treatment response. MDD is a leading cause of disability worldwide, and is commonly treated with AD. Although effective, many patients fail to respond to AD treatment, and accordingly identifying factors that can predict AD response would greatly improve treatment outcomes. In this study, we integrated multi-omic datasets (gene expression, DNA methylation, and genotyping) in order to identify biomarker profiles associated with AD response in a cohort of individuals with MDD. To address this rich multi-omic dataset with high dimensional features, we developed integrative Geneset-Embedded non-negative Matrix factorization (iGEM), which is a NMF based model supplemented with auxiliary information regarding genesets and gene-methylation relationships. Using our model, we identified a number of meta-phenotypes which were related to AD response. By integrating geneset information into our model, we were able to relate these meta-phenotypes to a number of biological processes, including immune and inflammatory functions, representing both potential new treatment targets as well as biomarkers to predict response. Our method is applicable to other diseases with multi-omic data, and the software is open source and available on Github (https://github.com/li-lab-mcgill/iGEM)
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 | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".