The nucleotide sequences, protein-coding sequences, protein sequences, chromosome location, strand, and sequence length were obtained based on the gene transfer format file (GTF). The physicochemical characteristics, including molecular weight (MW), theoretical point (pI), grand average of hydropathicity (GRAVY), and instability index (II), were calculated using the online tool ExPASy (http://web.expasy.org/protparam/).

A multiple sequence alignment of full-length proteins was performed using ClustalW v2.1. A neighbor-joining (NJ) tree was constructed using MEGA X with 1000 bootstrap replicates. An online multiple expectation maximization for motif elicitation (MEME) was used to detect conserved motif patterns with a maximum number of motifs set at 8 and an optimal motif width range from 6 to 50 amino acids. The intron and exon annotations were obtained from the GTF file and the gene structure was visualized using the Gene Structure Display Server (GSDS) (http://gsds.cbi.pku.edu.cn/). The 1.5 kb sequence, upstream, gene coding regions were extracted and submitted to the PlantCARE database to detect the cis-elements within promoters.

The protein sequences of Arabidopsis and rice were retrieved from the Ensembl Plants database (https://oct2017-plants.ensembl.org/index.html). Orthologous relationships between barley and Arabidopsis, and between barley and rice, were generated using the program Inparanoid v8.0. The synonymous substitution rate (Ks), non-synonymous substitution rate (Ka), and Ka/Ks ratio were estimated for orthologous gene pairs using codeml of PAML v4.3. The orthologous relationships were plotted using Circos v0.67. Divergence time was inferred using the formula T = Ks/2λ, where T is the time of duplication, Ks indicates the synonymous substitutions per site, and λ is the mutation rate of the divergence of plant nuclear genes (λ = 6.5 × 10^-9).

A total of 142 RNA-seq samples from different stages/tissues and adverse stresses were downloaded from the NCBI sequence reading archive (SRA) database. The accession number and BioProject information are as follows: PRJEB14349, RNA-Seq of 16 developmental stages of barley (Morex cultivar); PRJEB13621, a study of transcriptome response to salt treatment across three zones of the root in a barley landrace and malting cultivar; PRJEB18276, transcriptional responses to cold in five Pooideae species; PRJNA382490, the transcriptome of barley under three different heavy metal stress conditions. Low-quality reads were removed using Trimmomatic v0.39 (https://github.com/usadellab/Trimmomatic). Clean reads were aligned to the barley reference genome using HISAT v2.1.0. Aligned reads were sorted using SAMtools v1.3.1. The fragments per kilobase per million reads (FPKM) of each gene were calculated according to the reference annotation file. The expression level was visualized by the pheatmap package of R.(Sample_information.xlsx)

The whole-exome sequencing data of 220 geographically-referenced barley accessions were downloaded from the NCBI SRA database under the BioProject accession number: PRJEB8044 (Exome Capture to Study Genomic Diversity, Adaptation, and Selection in Barley). Read quality was evaluated and low-quality reads were filtered using Trimmomatic v0.39. The high-quality reads were mapped to the reference genome using BWA-MEM v0.7.13r1126. Picard v1.119 tools were used to clean, sort, and mark PCR duplicates of binary alignment map (BAM) files. Variant calling of BAM files was performed using the Haplotype Caller tool embedded in GATK v3.5-0-g36282e4. SNPs with minor allele frequency (MAF) <0.05 or >0.95, or missing rates >0.90 were removed. Only biallelic alleles were retained. Functional annotation and impact on coding regions of variation were determined using SnpEff v4.3. The upstream and downstream region of the gene was set to 3kb. (ERR_sample_information.xlsx)

The webserver was hosted on a lightweight application server of Tencent cloud (https://cloud.tencent.com/), freely accessible for non-commercial use via its website. Linux system CentOS v7.6.1810 (http://www.centos.org) was installed on the server. The front end of the webpage is implemented in HTML (https://www.w3.org/html/) CSS (http://www.w3.org) and JavaScript (https://www.javascript.com/), and PHP (https://www.php.net/) supported the server-side back-end. Multi-omics data were processed and stored in the MySQL v5.6.50 database server following the MySQL operations manual. JBrowse was installed to provide a user-friendly interface capable of accessing genome information of interest. ViroBLAST constructed a standalone database so online BLAST searches could be performed. Some local scripts were rewritten to provide additional search services.