Six Core Gene Markers for Colorectal Cancer Diagnosis

A diagnostic model based on six core genes (EIF2S3, GTF3A, HMGA1, HSP90AB1, PABPC1, S100A11) for identifying malignant transformation in colorectal polyps. The model was developed using bioinformatics and machine learning on the TCGA-COADREAD cohort and validated with external datasets. Research by Tao He demonstrates its diagnostic efficacy and association with patient prognosis.

Use Cases

• Validate the ridge regression diagnostic model's performance using the six core genes (EIF2S3, GTF3A, HMGA1, HSP90AB1, PABPC1, S100A11) on independent cohorts like GSE41258.
• Assess the correlation between high expression levels of EIF2S3 and S100A11 and poor prognosis in colorectal cancer patients via survival analysis.
• Verify upregulated protein expression of the six core genes in CRC tissues using data from the UALCAN database.
• Compare mRNA expression levels of the six genes between CRC cell lines (SW480, HCT116) and the normal cell line NCM460 using qRT-PCR data.

Strengths

• Model built on the TCGA-COADREAD cohort, a major public cancer genomics resource.
• Diagnostic performance validated in an external independent cohort (GSE41258).
• Findings supported by multiple validation methods: bioinformatics, survival analysis, and laboratory qRT-PCR.
• Dataset is openly licensed under CC BY 4.0 for reuse.

Limitations

• The dataset is very small (275.5 KB), containing only the final model's gene markers and related results, not the raw genomic data.
• Input data specifics like row count, column names, and sample data are unavailable, limiting direct analytical utility.
• Scope is restricted to the six identified genes; broader genomic or clinical context is not included in this file.

Provenance

Source

Analysis derived from public datasets GSE209741, GSE161277, and the TCGA-COADREAD cohort.

Collection Method

Bioinformatics analysis using edgeR and WGCNA, combined with machine learning feature selection (Boruta, LASSO, XGBoost) and ridge regression modeling.

Freshness

Last updated March 24, 2026.