We evaluated the performance of SEQMINER for annotating and querying summary association statistics using datasets from the 1000 Genomes phase 1 project. The call set consisted of 1,092 individuals genotyped at ∼39 million variants. We simulated phenotypes under the null hypothesis of no genotype‐phenotype associations. Summary association statistics were generated using RVTESTS [https://github.com/zhanxw/rvtests]. The output from RVTESTS is automatically bgzip‐compressed and tabix‐indexed. The file for single variant association statistics after compression is 6.2 GB, and the file for the covariance matrix between single variant score statistics after compression is 266 GB.
