Fig. 2

Sequence properties of anchored regions mediating inter-chromosomal interactions. a, b For randomly sampled read pairs, the frequency distributions of all chromosomal interactions without (a) or with (b) multi-mapped reads. Read pairs sampled from CS re-sequencing data [30]. c, d Distributions of sequence homology between anchored region pairs of significant inter-chromosomal interactions (inter-chr) within or between subgenomes, as well as anchored region pairs of significant intra-chromosomal interactions (intra-chr). Default BLAST settings were used to compare the sequence homology of read pairs. The proportions of aligned regions are presented. e Number of significant inter-chromosomal interactions within or between subgenomes with only uniquely mapped reads or with multi-mapped reads included. f Proportions of non-TEs and different types of TEs in the anchored regions supporting different types of chromosomal interactions. Significant inter-chromosome interactions within or between subgenomes were used. g Ternary plot presenting the relative abundance of 528 TE families in AK58 following the method published previously [31]. Each circle represents a TE subfamily within subgenome A, B or D; the coordinate represents the abundance of each TE subfamily in each subgenome relative to the abundance of the TE subfamily in all subgenomes. The TE subfamilies close to vertices correspond to single subgenome-dominant categories, whereas the TE subfamilies close to the edges and between vertices correspond to subgenome-suppressed categories. Balanced triads are indicated in gray. h Proportions of subgenome-biased TEs in significant inter-chromosomal interacting regions with a high sequence similarity (> 40% of the region alignable between anchored pairs). The distribution of within- and between-subgenome (subg) interacting pairs, as well as the random pairs, is shown