IAnimal

IAnimal integrates cross-species multi-omics datasets, literature-mined gene and trait entities, and genome annotations to support analysis of gene regulation and trait associations.

Key Features:

• Extensive Data Repository: Hosts 61,191 individual omics datasets covering genome (Whole Genome Sequencing - WGS), transcriptome (RNA-Seq), epigenome (ChIP-Seq, ATAC-Seq), and genome annotation information.
• Multi-species Coverage: Includes data from 21 species, explicitly listing mice, pigs, cattle, chickens, and macaques.
• Data Volume: Aggregates a total clean data volume of 846.46 terabytes.
• Deep Learning Integration: Employs BioBERT and AutoNER to extract "gene" and "trait" entities from 2,794,237 abstracts.
• Entity Extraction from Literature: Extracts gene and trait entities at scale to link omics signals with biological traits.
• Visualization Capabilities: Provides computational tools for visualizing complex multi-omics data.

Scientific Applications:

• Gene Regulation Studies: Enables exploration of how genome, transcriptome, and epigenome layers contribute to gene regulation.
• Trait Analysis: Supports linking genes to specific traits to investigate the genetic basis of phenotypic variation.
• Cross-Species Comparisons: Facilitates comparative analyses across 21 species to identify conserved processes and evolutionary patterns.

Methodology:

Data integration by aggregating diverse omics datasets; entity mining using BioBERT and AutoNER on 2,794,237 abstracts to extract gene and trait entities; computational visualization of multi-omics data.