Journal: Cancers

Article Title: Loss of CHGA Protein as a Potential Biomarker for Colon Cancer Diagnosis: A Study on Biomarker Discovery by Machine Learning and Confirmation by Immunohistochemistry in Colorectal Cancer Tissue Microarrays

doi: 10.3390/cancers14112664

Figure Lengend Snippet: Schematic flow chart of the present study for early diagnosis of colon adenocarcinoma. The starting materials were derived from the RNA seq data in the TCGA and GTEx databases. Differential expressed analysis (DEA) between the colon cancer patients and normal controls was conducted. The differential expressed (DE) genes were then mapped to the Human PPI network (from String) to construct a colon cancer-specific PPI network, and machine learning was used to predict new potential biomarkers based on the network features of the confirmed biomarkers from our CBD database. The diagnostic test (ROC test) of the predicted biomarkers were further verified in GEO microarray data. The candidate biomarker (CHGA) was finally confirmed by immunohistochemistry tissue microarrays.

Article Snippet: Then, after blocking the activity of endogenous peroxidase, the sections were incubated with the CHGA monoclonal rabbit anti-human IgG (CM10C, BIOCARE MEDICAL) in a 1:100 dilution with antibody dilution buffer overnight.

Techniques: Biomarker Discovery, Derivative Assay, RNA Sequencing, Construct, Diagnostic Assay, Microarray, Immunohistochemistry

Journal: Cancers

Article Title: Loss of CHGA Protein as a Potential Biomarker for Colon Cancer Diagnosis: A Study on Biomarker Discovery by Machine Learning and Confirmation by Immunohistochemistry in Colorectal Cancer Tissue Microarrays

doi: 10.3390/cancers14112664

Figure Lengend Snippet: PPI network and biological function analysis of the predicted biomarkers. ( A ) PPI relationships and enriched pathways (table) for the 12 predicted biomarkers. A total of seven predicted biomarkers related to each other. In addition, nine predicted biomarkers were mapped on the response to stress pathway. The candidate biomarkers that had a strong relationship were mapped in the same pathway. ( B ) PPI network for predicted and confirmed biomarkers. The color of lines represented the confidence of connected evidence: the closer to red, the higher evidence. Generally, the predicted and confirmed biomarkers had strong relationships with each other; specifically, not as the other ten predicted, which are connected closely, CHGA and SDC2 were separated from them and were hubs in their belonged small networks. ( C ) KEGG pathway enrichment analysis results for predicted and confirmed biomarkers. Big circles represent enriched pathways, and small circles/diamonds represent confirmed/predicted biomarkers. Pathways and biomarkers were connected if the biomarkers were mapped on the pathways. There were some overlapping pathways among the confirmed and predicted biomarkers, and five of them were mapped on the Dopaminergic synapse pathway. ( D ) Dopaminergic synapse pathway. ( E ) Diagnostic ROC curve for multiple biomarkers combined by the 12 predicted biomarkers, which showed an AUC of 0.964.

Article Snippet: Then, after blocking the activity of endogenous peroxidase, the sections were incubated with the CHGA monoclonal rabbit anti-human IgG (CM10C, BIOCARE MEDICAL) in a 1:100 dilution with antibody dilution buffer overnight.

Techniques: Diagnostic Assay

Journal: Cancers

Article Title: Loss of CHGA Protein as a Potential Biomarker for Colon Cancer Diagnosis: A Study on Biomarker Discovery by Machine Learning and Confirmation by Immunohistochemistry in Colorectal Cancer Tissue Microarrays

doi: 10.3390/cancers14112664

Figure Lengend Snippet: Protein expression of CHGA in normal mucosa, normal adjacent mucosa, and tumor from the same patient with colon cancer. The CHGA protein was positively expressed in the normal, and adjacent mucosa (the brown colour) and absolutely lost the CHGA expression in the tumor regardless of well-, moderately-, or poorly-differentiated cancers. Magnifications 10× and 40×.

Article Snippet: Then, after blocking the activity of endogenous peroxidase, the sections were incubated with the CHGA monoclonal rabbit anti-human IgG (CM10C, BIOCARE MEDICAL) in a 1:100 dilution with antibody dilution buffer overnight.

Techniques: Expressing

Journal: Cancers

Article Title: Loss of CHGA Protein as a Potential Biomarker for Colon Cancer Diagnosis: A Study on Biomarker Discovery by Machine Learning and Confirmation by Immunohistochemistry in Colorectal Cancer Tissue Microarrays

doi: 10.3390/cancers14112664

Figure Lengend Snippet: CHGA expression distribution of colon cancer ( A ), rectal cancer ( D ) and CRC ( G ) compared with normal controls. CHGA had a significant difference in cancers with normal controls. Diagnostic ROC test for CHGA in colon cancer ( B ), rectal cancer ( E ), and CRC ( H ). With AUCs of 0.995, GHGA showed a high potential of being a good diagnostic biomarker in CRC. Survival curves of CHGA in colon cancer patients ( C ), rectal cancer patients ( F ), and CRC patients ( I ). CHGA performed poorly in the prognosis of CRC ( p value = 0.24, 0.38, and 0.13, respectively).

Article Snippet: Then, after blocking the activity of endogenous peroxidase, the sections were incubated with the CHGA monoclonal rabbit anti-human IgG (CM10C, BIOCARE MEDICAL) in a 1:100 dilution with antibody dilution buffer overnight.

Techniques: Expressing, Diagnostic Assay, Biomarker Discovery