Skip to main content

Table 5 –blocks runtimes (sec)

From: Second-generation PLINK: rising to the challenge of larger and richer datasets

Parameters Dataset Machine Haploview 4.2 PLINK 1.07 PLINK 1.90
–ld-window-kb 500 synth1 Mac-2 nomem 3198.4 1.7
   Mac-12 45 k 3873.0 1.3
   Linux32-2 nomem 5441.1 3.4
   Linux64-512 57 k 2323.4 2.9
   Win32-2 nomem 9803.4 8.9
   Win64-2 51 k 5513.4 2.8
–ld-window-kb 1000 synth1 Mac-2 nomem 6185.7 2.2
   Mac-12 45 k 7394.4 9.8
   Linux32-2 nomem 9876.8 10.0
   Linux64-512 57 k 4462.1 3.9
   Win32-2 nomem 18925.7 17.3
   Win64-2 51 k 10.3 k 3.6
–ld-window-kb 500 chr1 Mac-2 nomem 2700 k? 550.9
   Mac-12 nomem 3600 k? 426.0
   Linux32-2 nomem 4300 k? 1288.4
   Linux64-512 440 k? 2600 k? 1119.7
   Win32-2 nomem 17000 k? 4535.8
   Win64-2 nomem 5700 k? 1037.2
  1. This command breaks the genome into estimated “haplotype blocks” which are usually inherited together. The PLINK 1.9 implementation combines optimizations recently developed by Taliun et al. [26] with additional laziness and bit-level parallelism.