Skip to main content
. 2019 Mar 21;9:4971. doi: 10.1038/s41598-019-41369-w

Table 3.

Topology consistency check of the models produced by Hierarchical3DGenome and LorDG without domain partitioning.

Hierarchical3DGenome LorDG Without Partition
1 0.898720584 0.074093704
2 0.933686968 0.025649295
3 0.938667819 0.074893363
4 0.988076596 0.013585756
5 0.875989854 −0.020093576
6 0.911444415 0.230877171
7 0.72885098 0.096554188
8 0.962162304 0.0460202
9 0.911241704 0.27890049
10 0.908920242 0.100821567
11 0.856633536 0.127690704
12 0.933434252 0.113935156
13 0.91778003 0.181649622
14 0.8897597 0.279229499
15 0.851429745 0.06654513
16 0.900103388 0.011566887
17 0.914467918 0.031162663
18 0.899153126 0.094189398
19 0.810298639 0.400262366
20 0.823568359 0.202652796
21 0.757115911 0.37725465
22 0.751419102 0.445408394

The Spearman’s correlation between distances from models at 1 MB resolution and those at 5 KB resolution of Hierarchical3DGenome and LorDG without domain partitioning. Chromosomes models generated by Hierarchical3DGenome is much more consistent with models at 1 MB resolution that the models generated by LorDG without domain partioning.