genereg

genereg constructs predictive linear models to quantify regulatory relationships among gene expression, promoter methylation, and transcription factor interactions in cancer to identify key regulatory elements and pathway-specific associations.

Key Features:

• Quantitative modeling: Constructs predictive linear models per target gene using gene expression, promoter methylation, and candidate transcription factor interactions.
• Data integration: Integrates ENCODE and TCGA datasets to inform model predictors.
• Pathway-specific analysis: Applied to Ovarian Serous Cystadenocarcinoma with a focus on DNA REPAIR, STEM CELLS, and GLUCOSE METABOLISM pathways.
• Cross-gene predictors: Considers expression of genes within the same or different pathways as model predictors.
• Model validation and experimental support: Includes comparative validation using ARACNe in basal-like Breast Cancer and supports experimental confirmation of predictions.

Scientific Applications:

• Model reliability and validation: Enables validation of predictive regulatory models across cancer types, including comparison with ARACNe results in basal-like Breast Cancer.
• Pathway investigation in ovarian cancer: Facilitates identification of pathway-specific regulatory relationships in Ovarian Serous Cystadenocarcinoma (DNA REPAIR, STEM CELLS, GLUCOSE METABOLISM).
• Discovery of regulatory elements and targets: Aids identification of known and novel regulatory elements, gene correlations, and potential pharmacological or clinical targets.
• Experimental follow-up: Supports experimental confirmation of computationally predicted gene regulatory relationships.

Methodology:

Constructs predictive linear models for each target gene using gene expression levels, promoter methylation status, expression of genes within the same or different pathways, and potential transcription factor interactions, integrating ENCODE and TCGA data and using ARACNe for comparative validation.