# Table 3

Posterior distribution of ratio of different modes’ locations for MSR5p (if existing)
Posterior of ratio of modes locations
Mode Intervala Mode Intervala Probabilityb Mean Median 95% p.i.c 98⅓% p.i.
ASN 2 [40,48] 1 [21,29] 1.00 1.80 1.80 [1.65,1.96] [1.63,2.00]
3 [60,86] 1 [21,29] 0.97 2.92 2.92 [2.60,3.26] [2.54,3.35]
3 [60,86] 2 [40,48] 0.97 1.62 1.62 [1.50,1.75] [1.47,1.80]
CEU 2 [48,59] 1 [9,40] 0.77 2.24 2.13 [1.56,3.47] [1.47,3.92]
3 [65,86] 1 [9,40] 0.53 3.25 3.12 [2.21,5.00] [2.09,5.57]
3 [65,86] 2 [48,59] 0.68 1.45 1.45 [1.36,1.55] [1.34,1.57]
YRI 2 [46,65] 1 [19,45] 0.88 1.67 1.66 [1.44,1.96] [1.40,2.08]
3 [72,92] 1 [19,45] 0.88 2.49 2.47 [2.08,3.00] [2.00,3.20]
3 [72,92] 2 [46,65] 1.00 1.49 1.49 [1.38,1.60] [1.35,1.62]
Total 2 [41,59] 1 [21,30] 0.93 1.91 1.92 [1.74,2.04] [1.70,2.08]
3 [68,84] 1 [21,30] 0.85 3.01 3.00 [2.71,3.25] [2.67,3.29]
3 [68,84] 2 [41,59] 0.91 1.57 1.57 [1.49,1.67] [1.46,1.70]

ainterval of repeat units where the corresponding mode should be located, which directly results from the data and is visualized in the bottom panels of Figure 4. These are the intervals with posterior probability of a (non-minor) mode, whereas between these intervals the posterior probability of a mode is negligible. Our findings on the modes’ locations therefore are derived from the data, not from a personal choice of intervals.

bposterior probability that both modes in the two given intervals exist.

cposterior interval (i.e. Bayesian counterpart of confidence interval).

Schaap et al.

Schaap et al. BMC Genomics 2013 14:143   doi:10.1186/1471-2164-14-143