Six Core Gene Diagnostic Model for Colorectal Cancer

This dataset contains six core genes (EIF2S3, GTF3A, HMGA1, HSP90AB1, PABPC1, S100A11) identified as malignant transformation markers for colorectal cancer. It was developed by Tao He in 2026 using bioinformatics and machine learning analysis of datasets including GSE209741, GSE161277, and TCGA-COADREAD. The diagnostic model was validated on internal and external cohorts, including GSE41258.

Use Cases

• Validate the six core gene diagnostic model (EIF2S3, GTF3A, HMGA1, HSP90AB1, PABPC1, S100A11) against new patient expression data.
• Investigate the correlation between high expression of EIF2S3 and S100A11 genes and poor prognosis in colorectal cancer patients.
• Use the identified gene markers to build a ridge regression classifier for distinguishing colorectal polyps from malignant tumors.

Strengths

• Model validation was performed on an external independent cohort (GSE41258).
• Six specific core genes are identified and listed, providing clear targets for analysis.
• Protein expression levels for the core genes were validated using the UALCAN database.

Limitations

• The dataset is very small (10.2 KB), containing only the model's core gene list and likely summary results, not raw expression data.
• The underlying data sources (GSE209741, TCGA-COADREAD) are not included, limiting reproducibility and direct analysis.
• The row count and column structure of the provided XLSX file are unknown.

Provenance

Source

Derived from public datasets GSE209741, GSE161277, TCGA-COADREAD, and validated on GSE41258.

Collection Method

Bioinformatics analysis using edgeR, WGCNA, and machine learning feature selection (Boruta, LASSO, XGBoost) to identify genes, followed by ridge regression modeling.

Freshness

Last updated March 24, 2026.