GEPAS

GEPAS performs comprehensive analysis of microarray data, including normalization, differential gene expression, clustering, class prediction, array comparative genomic hybridization (aCGH) analysis, and functional annotation to support biological and clinical interpretation.

Key Features:

• Data pre-processing: Provides normalization options for Affymetrix and two-color microarray experiments.
• Differential gene expression analysis: Implements algorithms to identify genes differentially expressed among classes or correlated with clinical outcomes, supporting time-course and dose-response designs.
• Clustering methods: Includes both established and novel clustering techniques for analysing gene co-expression across conditions.
• Class prediction and gene selection: Supports development of predictive models and selection of genes relevant to classification and clinical outcomes.
• Array Comparative Genomic Hybridization (aCGH): Contains modules such as InSilicoCGH for managing and analysing aCGH data and mapping expression and genomic data onto chromosomes.
• Functional annotation: Integrates functional profiling using Gene Ontology, pathways databases, PubMed abstracts, transcription factor binding sites, chromosomal locations, and tissue-specific information.
• Automation (Pipeliner module): Provides a pipeliner module to automate sequential analysis steps.
• Extended gene identifier database: Includes an extended database of gene identifiers for identifier handling.

Scientific Applications:

• Basic biological research: Analysis of gene expression patterns and co-expression structure to generate biological hypotheses.
• Clinical studies: Identification of genes associated with clinical outcomes and development of predictive classifiers.
• Time-course and dose-response experiments: Analysis of temporal and dose-related expression changes using dedicated differential expression methods.
• Integration of expression and genomic data: Mapping and joint analysis of expression and aCGH genomic alterations on chromosomes.
• Functional interpretation and pathway analysis: Linking microarray results to Gene Ontology, pathway databases, PubMed abstracts, transcription factor binding sites, chromosomal locations, and tissue-specific information.

Methodology:

Computational methods explicitly include normalization for Affymetrix and two-color microarrays; differential expression algorithms for class comparisons and correlations with clinical outcomes including time-course and dose-response designs; established and novel clustering algorithms; class prediction and gene selection procedures; aCGH analysis with InSilicoCGH for mapping expression and genomic data onto chromosomes; functional profiling using Gene Ontology, pathways databases, PubMed abstracts, transcription factor binding sites, chromosomal locations, and tissue-specific information; and automation of sequential analysis steps via the Pipeliner module.