Accurate data capture is essential to make data correctly available for searching, visualization and integration with other resources. To ensure the GWAS Catalog is able to accurately capture study design and association results of increasing complexity we reviewed and redesigned the Catalog database schema. The representation of effect sizes has been improved with odds ratio and beta coefficient information now captured and displayed in separate fields, improving the accessibility of this data for users. We also mandated the use of structured, rather than free text, terms for key concepts including beta unit and direction, platform manufacturer, number of SNPs analyzed and whether imputation was used. This will improve consistency and integration potential with other complementary resources. Changes were made to the Catalog infrastructure to improve the representation of composite genomic elements, including haplotypes and SNP-by-SNP associations. Each individual variant is now captured separately in the database allowing variant-specific mapping, searches, links and visualization. The new schema design is also more flexible and will support future extensions to the scope of the GWAS Catalog to meet evolving study designs and users needs, as discussed in future work.
