R400: A novel gene signature for dose prediction in radiation exposure studies in humans
Bibliographic record
Abstract
Radiation’s harmful effects on biological organisms have long been studied through mainly evaluating pathological changes in cells, tissues, or organs. Recently, there have been more accessible gene expression datasets relating to radiation exposure studies. This provides an opportunity to analyze responses at the molecular level toward revealing phenotypic differences. Biomarkers in toxicogenomics have been suggested as indicators of radiation exposure and seem to react differently to various dosages of radiation. This study proposes a predictive gene signature specific to radiation exposure and can be used in automatically diagnosing the exposure dose. In searching for a reliable gene set that will correctly identify the exposure dose, consideration needs to be given to the size of the set. For this reason, we experimented with the number of genes used for training and testing. Gene set sizes of 28, 100, 200, 300, 400, 500, 600, 700, 800, 900 and 1,000 were tested to find the size that provided the best accuracy across three datasets. Models were then trained and tested using multiple datasets in various ways, including an external validation. The dissimilarities between these datasets provide an analogy to real-world conditions where data from multiple sources are likely to have variances in format, settings, time parameters, participants, processes, and machine tolerances, so a robust training dataset from many heterogeneous samples should provide better predictability. All three datasets showed positive results with the correct classification of the radiation exposure dose. The average accuracy of all three models was 88% for gene sets of both 400 and 1,000 genes. R400 provided the best results when testing the three datasets used in this study. A literature validation of top selected genes shows high relevance of perturbations to adverse effects reported during cancer radiotherapy.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".