CamurWeb: a classification software and a large knowledge base for gene expression data of cancer

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Next Generation Sequencing data necessitates advanced knowledge extraction methods.
• RNA sequencing gene expression is crucial for cancer case-control studies.
• Existing methods extract limited gene-based models, hindering comprehensive cancer insight.

Purpose of the Study:

• To develop a novel method and web-based software, CamurWeb, for extracting multiple, equivalent rule-based classification models.
• To create a comprehensive, queryable knowledge base of gene associations with tumors.
• To facilitate the discovery of genes potentially involved in cancer.

Main Methods:

• An iterative classification procedure combined with adaptive feature elimination.
• Development of CamurWeb software with a user-friendly interface for data upload, analysis, and result interpretation.
• Application to all publicly available RNA sequencing datasets from The Cancer Genome Atlas (TCGA).

Main Results:

• CamurWeb successfully extracted numerous rule-based models from TCGA RNA sequencing data.
• A large, open-access knowledge base of gene expression patterns across 21 cancer types was generated.
• The software demonstrated its capability to identify multiple genes associated with specific cancers.

Conclusions:

• CamurWeb validates its approach by generating extensive classification models and associated genes for diverse cancers.
• The comprehensive knowledge base and software are publicly accessible for further cancer research.
• This resource supports the design of biological experiments and advances understanding of cancer genetics.