Group and Basis Restricted Non-Negative Matrix Factorization and Random Forest for Molecular Histotype Classification and Raman Biomarker Monitoring in Breast Cancer
Bibliographic record
Abstract
Raman spectroscopy is a non-invasive optical technique that can be used to investigate biochemical information embedded in cells and tissues exposed to ionizing radiation used in cancer therapy. Raman spectroscopy could potentially be incorporated in personalized radiation treatment design as a tool to monitor radiation response in at the metabolic level. However, tracking biochemical dynamics remains challenging for Raman spectroscopy. Here we developed a novel analytical framework by combining group and basis restricted non-negative matrix factorization and random forest (GBR-NMF-RF). This framework can monitor radiation response profiles in different molecular histotypes and biochemical dynamics in irradiated breast cancer cells. Five subtypes of; human breast cancer (MCF-7, BT-474, MDA-MB-230, and SK-BR-3) and normal cells derived from human breast tissue (MCF10A) which had been exposed to ionizing radiation were tested in this framework. Reference Raman spectra of 20 biochemicals were collected and used as the constrained Raman biomarkers in the GBR-NMF-RF framework. We obtained scores for individual biochemicals corresponding to the contribution of each Raman reference spectrum to each spectrum obtained from the five cell types. A random forest classifier was then fitted to the chemical scores for performing molecular histotype classifications (HER2, PR, ER, Ki67, and cancer versus non-cancer) and assessing the importance of the Raman biochemical basis spectra for each classification test. Overall, the GBR-NMF-RF framework yields classification results with high accuracy (>97%), high sensitivity (>97%), and high specificity (>97%). Variable importance calculated in the random forest model indicated high contributions from glycogen and lipids (cholesterol, phosphatidylserine, and stearic acid) in molecular histotype classifications.
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 | 0.003 |
| 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 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| 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".