Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2020 Jan 24;2(1):lqaa003. doi: 10.1093/nargab/lqaa003

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© The Author(s) 2019. Published by Oxford University Press on behalf of NAR Genomics and Bioinformatics.

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com

PMC Copyright notice

Figure 3. — Resource usage and computation time. (A) The distribution of the number of SNPs per region when using the recombination hotspot map proposed by Berisa and Pickrell (27) while using 1M SNPs from the Illumina Omni1-QuAd, 9.9M SNPs from a recent GWAS and 96M SNPs as available in the latest UK Biobank genotype panel after imputation. (B) The trade-off between memory usage and computation time. JASS omnibus statistics was computed for 4 traits and 3 192 045 SNPs (insulin) and 11 traits and 3 673 285 SNPs (metabolism). The y-axis is the peak in memory usage in megabytes and x-axis is the time (in seconds) to perform the task. Labels attached to data points give the number of genomic regions processed at once. For example the two boxes filled with a “2000" in the top left of the plot correspond to the case were the entire dataset is loaded in memory (as there is a total of 1703 regions in this example). Color represents the group of traits. (C) The computation time optimization. We computed the omnibus statistics for 3 673 285 SNPs for 11 traits; the y-axis is the time in seconds (logarithmic scale) and x-axis is the number of missing data in a pattern. The straight green line shows the result from a naïve computation loop approach over all SNPs, and the red line shows the results from our optimized algorithm.