GOSim

GOSim computes information-theoretic functional similarity measures between genes using Gene Ontology (GO) annotations to quantify functional relatedness and support clustering and homogeneity assessment of gene sets from high-throughput data such as DNA microarrays.

Key Features:

• Functional Similarity Calculation: Computes pairwise and groupwise functional similarity between gene products using information-theoretic measures derived from GO terms.
• Gene Clustering by Biological Function: Clusters genes based on GO-derived similarity metrics to group genes with shared functional characteristics.
• Evaluation of Gene Group Homogeneity: Quantifies the homogeneity of gene sets with respect to their GO annotations to assess functional coherence.
• Complementary Analysis to GO Enrichment: Provides functional similarity–based insights that complement overrepresentation and enrichment analyses of GO terms.

Scientific Applications:

• Gene Expression Studies: Integrates GO annotations with expression data to interpret functional relationships among differentially expressed genes.
• Functional Genomics: Supports analysis of gene function and interaction networks by quantifying functional relatedness among gene products.
• Data Integration: Combines experimental datasets with Gene Ontology knowledge to enhance interpretation of genomic analyses.

Methodology:

Implemented in the R programming language, GOSim employs information-theoretic concepts to derive similarity measures between GO terms based on the GO hierarchical structure (molecular function, biological process, cellular component) to quantify functional relatedness of genes.