TWAS Atlas: A Knowledgebase of Transcriptome-wide Association Studies

The rising of transcriptome-wide association studies (TWAS) has provided an efficient analysis strategy to detect key genes of complex traits or diseases. In recent years, a large number of TWAS researches were conducted and accumulated a huge volume of analysis data, which laid the foundation for further analysis such as data integration and visualization.

To fill the gap, the research group from the Beijing Institute of Genomics of Chinese Academy of Sciences/ China National Center for Bioinformation (CNCB) developed TWAS Atlas, a knowledgebase of transcriptome-wide association studies, with the aim of exploring and mining gene-trait associations.The work was published in Nucleic Acids Research with the title “TWAS Atlas: a curated knowledgebase of transcriptome-wide association studies”.

TWAS Atlas offers multiple ways for users to browse, search and download integrated data. Until now, TWAS Atlas 1.0 version has integrated high-quality TWAS data from 200 TWAS publications via manual curation, covering 401,266 gene-trait associations, 257 traits, 22,247 genes and 135 tissue types for human. TWAS Atlas also consists of the meta-data for each research, showing the source and annotations of TWAS data.

In addition, TWAS Atlas provides an online tool to construct a comprehensive and interactive knowledge graph for SNP-gene-trait relationships through systematically integrating gene-trait association from TWAS and SNP-trait regulatory information from the GTEx database. The knowledge graph enables users to perform real-time interpretation and visualization of systematic network analysis incorporating multiple diseases, tissues and multi-omics data, providing personalized reference to explore multi-layer regulations on traits.

An application example of knowledge graph (center node set as (A) epithelial ovarian cancer, (B) epithelial ovarian cancer-related gene MAPT and (C) shared gene of epithelial ovarian cancer and Parkinson's disease) (Image by XIAO Jingfa’s group)

Contact:

Dr. XIAO Jingfa
Email: xiaojingfa@big.ac.cn