Additional file 1 of Transcriptional patterns of cancer-related genes in primary and metastatic tumours revealed by machine learning
Bibliographic record
Abstract
Additional file 1: This file contains a detailed description of the methodologies used for data analysis, along with supplementary tables S1-S62 and supplementary figures S1-S23. TableS1 – Tumour types selected for downstream analysis based on all samples. TableS2 – Tumour types selected for downstream analysis based on balanced sets of samples. Tables S3 to S17 – Number of mutant and wild-type samples and F1 scores based on alterations in ARID1A, BRAF, BRCA1, CDH1, CTNNB1, EGFR, EZH2, KDM6A, NRAS, PBRM1, PIK3CA, PTEN, SETBP1, SETD2, and SPOP. Tables S18 to S22 – Number of mutant and wild-type samples and F1 scores after excluding specific tumour types for ARID1A, EGFR, EZH2, PBRM1, and SPOP. Table S23 – Chromosomal regions excluded from analysis. Tables S24 to S56 – Top genes in classification of samples based on alterations in APC, ARID1A, ATR, ATRX, BRAF, BRCA1, CDH1, CDKN2A, CTCF, CTNNB1, EGFR, EZH2, FBXW7, GATA3, KDM6A, KEAP1, KIT, KRAS, MAP3K1, NCOR1, NF1, NOTCH1, NRAS, NSD1, PBRM1, PIK3CA, PTEN, RB1, SETBP1, SETD2, SF3B1, SPOP, and STAG2. Tables S57 to S59 – Top genes in classification of samples based on alterations in APC and BRAF under different settings. Table S60 – Top pathways affected by gene alterations. Table S61 – Top pathways affected by BRAF gene alterations in thyroid and colorectal cancers. Table S62 – Non-impactful mutations likely playing a role in pathogenesis. Figures S1 to S3 – PCA plots of POG, TCGA, and all samples. Figure S4 – Performance comparison across different models. Figure S5 – F1 scores based on different sets of gene alterations. Figure S6 – F1 score comparison between 5-fold CV and test set. Figure S7 – F1 scores distribution across all genes and tumour types. Figure S8 – Tumour-type-level F1 scores for BRAF. Figure S9 – F1 scores distribution based on balanced sets. Figure S10 – Tumour-type-level F1 scores for APC. Figures S11 to S14 – Gini scores based on true and randomly shuffled labels for KRAS, PTEN, AR, and ERBB4. Figure S15 – F1 score distribution across all genes. Figure S16 – Gini score distribution across all genes. Figures S17 to S19 – Top genes SHAP values for ATRX, BRAF and TP53. Figures S20 to S23 – Samples with intron variants predicted as mutant for EGFR, EZH2, NCOR1, and RB1.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.072 | 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 teacher head, 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".