Rummagene

Rummagene mines supplementary materials from PubMed Central to extract and aggregate mammalian gene sets (642,389 sets identified in 121,237 articles from over 5.4 million scanned articles) for downstream enrichment, similarity, and gene function analyses including transcription factor and kinase enrichment.

Key Features:

• Massive Data Mining: Systematically crawls PubMed Central to extract gene sets from supporting materials, having scanned over 5.4 million articles and identified 121,237 articles containing 642,389 mammalian gene sets.
• Enrichment Analysis: Performs transcription factor and kinase enrichment analyses to assess regulatory and signaling associations.
• Search Functionality: Supports free-text and table-title searches to locate gene sets within supplementary materials.
• Gene Set Similarity and Abstract Analysis: Integrates gene set similarity with abstract similarity analysis to identify relationships between biological processes, concepts, and named entities.
• Gene Function Prediction: Predicts gene functions by leveraging statistical patterns within the aggregated gene-set database.

Scientific Applications:

• Transcription Factor and Kinase Research: Enables identification of candidate transcription factors and kinases through enrichment analyses of published gene sets.
• Gene Function Studies: Supports hypothesis generation for gene function by providing predicted associations derived from aggregated gene-set patterns.
• Discovery of Novel Relationships: Facilitates discovery of previously unrecognized connections between genes, biological concepts, and named entities via combined gene-set and abstract similarity analyses.

Methodology:

Systematic crawling of PubMed Central to extract gene sets from supporting materials; transcription factor and kinase enrichment analyses; free-text and table-title searches of extracted tables; computation of gene-set similarity and abstract similarity analyses; and gene function prediction using statistical patterns in the aggregated database.