*2.1. Plant Materials*

Common wheat cv. Chinese Spring (CS), *Thinopyrum intermedium* (Host) Barkworth & D.R. Dewey, and *Th. ponticum* (Podp.) Z.-W.Liu & R.-C.Wang were provided by Prof. Bikram S. Gill, Wheat Genetic and Genomic Resource Center, Kansas State University. Germplasm of *Aegilops* L. and *Hordeum* L. species were obtained from Prof. Eviatar Nevo at the Institute of Evolution, University of Haifa, Israel. The durum wheat—*Dasypyrum villosum* (L.) Borbás amphiploid was obtained from Prof. Huaren Jiang, Sichuan Agricultural University, China. Lines of tetraploid wheat, including *T. carthlicum* Nevski, *T. abyssinicum* L., *T. durum*, and *T. dicoccoides*, were provided by Dr. Cheng Liu, Crop Research Institute, Shandong Academy of Agricultural Sciences. Tetraploid *Ae. tauschii* (Zeng et al. 2012) [31], rye (*Secale cereale* L.), *D. villosum* L., hexaploid triticale cv. Currency, and the wheat—*D. breviaristatum* L. partial amphiploid [32] were maintained in our laboratory at the School of Life Science and Technology, University of Electronic Science and Technology of China. A list of accessions with their ploidy level and genome constitution are given in Table S1.

### *2.2. Identification of Tandem Repeats (TRs)*

The wheat genome IWGSC RefSeq v1.0 assembly and genomic annotation was downloaded from (https://wheat-urgi.versailles.inra.fr/). The genome-wide analysis of the tandem repeats in wheat was conducted by the Tandem Repeat Finder (TRF) algorithm [33] using alignment parameters of 2, 7, and 7 for match, mismatch, and indels, respectively. The interpretation of TRs by number of copy, array, and cluster is illustrated in Figure S1. Tandem repeats annotated by TRF were divided into three classes according to the size of period distances (<20, 20–60, and >60). A nonredundant set of sequences was constructed using the program CD-HIT [34] at an 85% similarity level. The distribution and chromosomally enriched locations of the clusters of TRs in wheat genome were visualized with the assistance of a website, B2DSC (http://mcgb.uestc.edu.cn/b2dsc), according to the procedure of Lang et al. [29].

### *2.3. Fluorescence In Situ Hybridization (FISH)*

Seedling roots from wheat and related species and wheat-alien derivatives were collected when they were about 2–3 cm long. Excised root tips were treated with nitrous oxide gas for 2 h under 1.0 MPa pressure. The treated root tips were fixed in 90% acetic acid for 10 min. After the roots were washed by distilled water, they were digested in 2% cellulase and 1% pectolyase (Yakult Pharmaceutical, Tokyo, Japan) for 55 min at 37 ◦C. The digested root sections were washed and then mashed to form a cellular suspension in 100% acetic acid. The cell suspensions were dropped onto glass slides for chromosome preparation according to Kato et al. [35]. The synthesized oligonucleotide probes Oligo-pSc119.2 with 5' labeled by 6-carboxyfluorescein (6-FAM) and Oligo-pTa535 with 5' labeled by 6-carboxytetramethylrhodamine (Tamra) were used for identifying the wheat chromosomes according to the description of Tang et al. [36]. The oligonucleotide probes were synthesized by Shanghai Invitrogen Biotechnology Co. Ltd. (Shanghai, China). The protocol of nondenaturing FISH (ND-FISH) employing synthesized probes was described by Fu et al. [37]. After ND-FISH, the chromosome squashes for sequential FISH for Oligo-3A1 labeled by 6-FAM were washed twice, each for 5 min with 0.1% Tween 20 in 2 × SSC, to remove the hybridization signals. FISH images were captured with an Olympus BX-51 microscope equipped with a DP-70 CCD camera operated by DP manager software (Olympus). Images were processed using Photoshop 3.0 (Adobe Systems Incorporated, San Jose, CA, USA).

### *2.4. Sequences Variation and Phylogenetic Analysis*

The DnaSP version 6.0 [38] was used to compute DNA polymorphism and their variances. Multiple sequence alignments from different chromosomal regional accumulated Oligo-3A1 sequences were done using ClustalX [39] and were subsequently used for phylogenetic analyses. Maximum likelihood (ML) trees were constructed by MEGA7.0 [40], and the trees were displayed with FigTree v1.40 software (http://tree.bio.ed.ac.uk/software/figtree/). Clustered consensus sequences combined with the phylogenetic trees were displayed with the logos [41,42].
