Extended Data Fig. 10: Runtime for LDGM and precision matrix inference across chromosome 22.
From: Extremely sparse models of linkage disequilibrium in ancestrally diverse association studies
The first five boxplots indicate the runtime for precision matrix inference for each ancestry group; the sixth indicates the runtime to derive the LDGM from the original tree sequence. For the LDGM inference step, we used 5 compute threads (1 for the precision matrix inference step). Runtime varies across LD blocks and ancestry groups due to variation in the number of SNPs. The lower whisker, lower hinge, center, upper hinge and upper whisker correspond to (lower hinge − 1.5× interquartile range (IQR)) and the 25th percentile, median, 75th percentile, and (upper hinge + 1.5× IQR), respectively. For numerical results, see Supplementary Table 4.
