Genotypic Data in Relational Databases: Efficient Storage and Rapid Retrieval

As technologies to produce genotypic data have become less expensive, the widths and depths of such data have sharply increased. Relational databases have performed poorly in this domain. Data storage and retrieval is now mostly conducted by highly coupled and specialized software packages and file formats, but relational databases offer advantages if the domain challenges can be overcome. We revisit their feasibility as a tool for efficiently storing and querying extremely large genotypic data sets. We describe a technique for managing genotypic data in the PostgreSQL relational database, compare it to common existing techniques for storing and querying genotypic data, and demonstrate that it can greatly reduce both query times and storage requirements.