YAGM

YAGM mines gene–gene associations in yeast by integrating transcription factor binding, transcriptional regulation, mutant phenotype, functional annotation, physical and genetic interaction, literature evidence, and expression-profile data to identify associated genes and networks.

Key Features:

• Comprehensive Data Integration: Integrates seven distinct datasets: Transcription Factor-Gene Binding (TFB), Transcription Factor-Gene Regulation (TFR), Mutant Phenotype (MP), Functional Annotation (FA), Physical Interaction (PI), Genetic Interaction (GI), and Literature Evidence (LE).
• Expression Profile Integration: Incorporates expression-profile association data sourced from the SPELL database.
• Novel Associations: Constructs eight biological association types in total, three of which are novel: TFB (transcription factor–gene binding), TFR (transcription factor–gene regulation), and MP (mutant phenotype).
• Configurable Association Combination: Supports generation of tailored associated-gene lists by combining any subset of the eight association types.
• Association Scoring: Calculates association scores between gene pairs across the integrated datasets.
• Distinctive Associated Gene Sets: Produces associated-gene sets that are distinct from those reported by STRING and GeneMANIA, providing alternative candidate genes for experimental validation.

Scientific Applications:

• Gene function and interaction inference: Identifies genes associated with a query yeast gene to support inference of gene function and interaction networks.
• Transcriptional regulation analysis: Facilitates analysis of transcription factor–gene binding and regulation relationships via TFB and TFR associations.
• Phenotype–genotype association: Enables linking of genes through mutant phenotype (MP) data to explore phenotype–genotype relationships.
• Candidate prioritization for validation: Provides alternative associated-gene candidates distinct from STRING and GeneMANIA for experimental follow-up.

Methodology:

Association scores between gene pairs are calculated across the integrated datasets using a hypergeometric test.