Genome synteny, Ks analysis, subgenome identification and homoeolog expression bias
Protein sequences within and between genomes were aligned using BLASTP with an e value cutoff of 1e-5. Syntenic blocks were detected based on homologous gene pairs using MCScanX (Wang et al., 2012) and MCScan (https://github.com/tanghaibao/jcvi/wiki/MCscan-[Python-version]).O. kokonorica and C. songorica are both tetraploids belonging to the same subtribe of Chloridoideae, and C. songoricahas been revealed to be an allotetraploid species with 20 chromosomes assigned to two subgenomes (Zhang et al., 2021a). Therefore, we partitioned the O. kokonorica genome into subgenomes A and B by dotplot with syntenic blocks and synonymous nucleotide substitutions (K s) values against the C. songorica subgenomes using WGDI (Sun et al., 2022). Analysis of homoeolog expression bias was preformed following VanBuren et al. (2020) for all five tissues and samples under three abiotic treatments (see below).
WGD and speciation events were inferred by calculating Ks values using MCScanX downstream analysis program (Perl). Tandem duplicated gene pairs were identified using duplicate_gene_classifier, a program of MCScanX. Functional classification of Gene Ontology (GO) categories was performed on the tandem duplicated genes using Blast2GO (Conesa et al., 2005). Fisher’s exact test was used to calculate the statistical significance of enrichment (p < 0.05) with Benjamini–Hochberg’s false discovery rate adjustment.