Prediction of Risk Factors from Gastric Cancer Genetic Data Using Machine Learning
Bibliographic record
Abstract
Conventional statistical methods are challenging to predict cancer risk factors due to complex, non-linear, interactions among genetic factors. They often fail to handle high-dimensional data and dynamic risk factors effectively. This paper aims to utilize machine learning techniques to identify key genetic features from genomic data that contribute to the development of gastric cancer. The dataset comprises 192,781 instances with 64 annotated genetic variant features from gastric cancer patients, subjected to thorough preprocessing and quality checks prior to analysis. Feature selection was employed to identify critical features, which are used to develop five ensemble classifiers: Bagging, Random Forest, Extra Trees, AdaBoost, and Gradient Boosting. The Extra Trees classifier achieved an accuracy of 97.57%, precision of 95.62%, recall of 90%, F1-score of 92.73%, and Matthews Correlation Coefficient (MCC) of 0.91 on the imbalanced dataset, with an Area Under the Curve (AUC) of 98% for the positive class. Random undersampling of the dataset yielded promising results, reinforcing the selected features' effectiveness with consensus classification, achieving an accuracy of 96.06%, precision of 95.74%, recall of 96.33%, F1-score of 96.04%, MCC of 0.92, and Receiver Operating Characteristic (ROC) of 96.06%. Feature selection applied to both balanced and imbalanced datasets markedly improved model performance metrics, enhancing interpretability and precision for the minority class. The extracted features substantially reduced computational complexity in next-generation sequencing analysis. Moreover, these features provide critical insights into identifying novel therapeutic targets by predicting interactions with disease-associated proteins, thereby facilitating molecular dynamics investigations. This methodology significantly advances the development of personalized medicine applications. The findings underscore the effectiveness of feature selection in optimizing genomic data analysis for precision healthcare.
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) | 0.000 | 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".