Deep Learning Based Clustering of Gene Expression Data in Cancer Patients using Variational Autoencoder Model
Bibliographic record
Abstract
Background: High mortality associated with cancer presents challenges in both diagnosis and treatment. Deep learning-based clustering analysis provides a powerful tool for identifying molecular subtypes within cancer enabling personalized treatment strategies. Objectives: This project has two objectives: (1) to cluster gene expression data for cancer patients into unique subtypes to uncover novel biological insights and improve patient outcomes; (2) to identify unique molecular signatures driving each cluster. Methods: GDC Pan-Cancer data, consisting of 33 tumor types with 11,506 samples and 32,967 gene expression features, from TCGA repository was used. Outliers were identified using interquartile criterion, and replaced using multiple imputation by chained equations. A variational autoencoder (VAE) model was used to extract latent representations within the gene expression data. These latent features were then utilized in K-Means clustering to identify distinct clusters based on similarity in gene expression profiles. Survival rate and median survival time for each cluster were computed, and statistical tests were used to determine the significance of the observed differences. Finally, gene signatures for individual clusters were developed using standardized mean difference (SMD) analysis, determining features significant in one cluster vs the others (p<0.05). Results: Survival rates vary significantly across clusters (p<0.05). Clustering helped identify clusters with low survival rates. Subsequently, distinct gene signatures identified for each cluster helped determine molecular mechanisms driving survival outcomes. Conclusion: A deep learning-based model efficiently groups cancer patients, revealing unique molecular signatures within each subgroup. Screening and prioritizing treatment for patients in subgroups with lower survival rates could enhance patient management.
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 | 0.001 |
| 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) | 0.001 | 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".