DIANE

DIANE infers and analyzes gene regulatory networks from multi-factorial expression datasets to identify regulator-target relationships and model gene expression patterns.

Key Features:

• Comprehensive Workflow: Provides steps including normalization, dimensionality reduction, differential expression analysis, and ontology enrichment for raw RNA-seq datasets.
• Advanced Gene Clustering: Performs model-based gene clustering using configurable Mixture Models to explore expression patterns across conditions.
• Gene Regulatory Network Inference: Infers gene regulatory networks using Random Forests to estimate regulator-target relationships.
• Statistical Significance Assessment: Assesses significance of regulator-target influence measures via permutations of Random Forest importance metrics.
• Multi-factorial Dataset Support: Operates on multi-factorial expression datasets and can handle data from diverse organisms.

Scientific Applications:

• Exploring transcriptional responses to environmental perturbations: Applied to datasets such as Arabidopsis thaliana under combined temperature, drought, and salinity stresses to characterize transcriptomic responses.
• Identifying candidate genes and pathways: Enables exploratory analyses, model-based clustering, and network reconstruction to nominate candidate genes or signaling pathways for further investigation.

Methodology:

Computational methods explicitly include normalization, dimensionality reduction, differential expression analysis, ontology enrichment, model-based clustering with Mixture Models, network inference with Random Forests, and permutation testing of Random Forest importance metrics.