PRPS-ST: A Protocol-Agnostic Self-training Method for Gene Expression–Based Classification of Blood Cancers
Bibliographic record
Abstract
Abstract Gene expression classifiers are gaining increasing popularity for stratifying tumors into subgroups with distinct biological features. A fundamental limitation shared by current classifiers is the requirement for comparable training and testing datasets. Here, we describe a self-training implementation of our probability ratio-based classification prediction score method (PRPS-ST), which facilitates the porting of existing classification models to other gene expression datasets. In comparison with gold standards, we demonstrate favorable performance of PRPS-ST in gene expression–based classification of diffuse large B-cell lymphoma (DLBCL) and B-lineage acute lymphoblastic leukemia (B-ALL) using a diverse variety of gene expression data types and preprocessing methods, including in classifications with a high degree of class imbalance. Tumors classified by our method were significantly enriched for prototypical genetic features of their respective subgroups. Interestingly, this included cases that were unclassifiable by established methods, implying the potential enhanced sensitivity of PRPS-ST. Significance: The adoption of binary classifiers such as cell of origin (COO) has been thwarted, in part, by the challenges imposed by batch effects and continual evolution of gene expression technologies. PRPS-ST resolves this by enabling classifiers to be ported across platforms while retaining high accuracy. This article is highlighted in the In This Issue feature, p. 215
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 | 0.004 | 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 | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.003 |
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".