MIDAW

MIDAW analyzes microarray gene expression data using statistical routines to perform normalization, missing-value estimation, dimension reduction, discriminant analysis, marker-gene identification, and probe annotation to public sequence databases.

Key Features:

• Data Normalization: Implements background correction and global and local mean and variance normalization and supports simultaneous processing of multiple experiments.
• Descriptive Graphics: Provides graphical display options for descriptive analysis and visualization of expression patterns.
• Missing-Value Estimation: Includes estimation algorithms to impute missing values in microarray datasets.
• Dimension Reduction: Offers data dimension reduction techniques for handling large-scale microarray data.
• Discriminant Analysis: Performs discriminant analysis for classification of samples based on gene expression profiles.
• Marker Gene Identification: Identifies marker genes that serve as indicators for specific phenotypes or diseases.
• External Database Linking: Maps transcript/gene probes to GenBank, Entrez Gene, and UniGene.

Scientific Applications:

• Gene Expression Profiling: Supports normalization and statistical analysis of microarray gene expression datasets.
• Cancer Genomics: Facilitates identification of marker genes and expression signatures relevant to cancer research.
• Developmental Biology: Enables analysis of gene expression changes across developmental stages.
• Systems Biology: Aids exploration of gene regulation and interaction networks derived from microarray data.

Methodology:

Implemented using Perl and PHP with R/Bioconductor routines; explicitly includes background correction, global and local mean and variance normalization, simultaneous processing of multiple experiments, graphical displays, missing-value estimation, dimension reduction, discriminant analysis, marker-gene identification, and probe mapping to GenBank, Entrez Gene, and UniGene.