Skip to main content
. 2018 Dec 19;47(4):1605–1614. doi: 10.1093/nar/gky1263

Figure 6.

Figure 6.

Runtime comparison between WISECONDOR and WisecondorX. Minutes are presented in logarithmic scale. Timing covered the ‘bam to result’ principle (meaning the combination of the convert and test (WISECONDOR) or predict (WisecondorX) functions). Despite not being analyzed by this validation set, WISECONDOR exceeds a mean runtime of 24 h for resolutions of 15 kb (supposed exponential complexity for the test function), whereas WisecondorX barely increased with 2 min (supposed linear complexity for the predict function). The increased runtime in WISECONDOR is caused by the iterative Stouffer's z-score procedure.