Documentation of the analyses on genomic and protein data from the plants Lolium multiflorum and Sorghum bicolor on the weed resistance project for John.

Download the genome of Lolium multiflorum (Italina ryegrass) with the accession number GCA_019182485.1, unzip it and save it as in FASTA format named "GCA_019182485.1_Rabiosa_v.0_genomic.fasta".

wget https://ftp.ncbi.nlm.nih.gov/genomes/all/GCA/019/182/485/GCA_019182485.1_Rabiosa_v.0/GCA_019182485.1_Rabiosa_v.0_genomic.fna.gz

gunzip GCA_019182485.1_Rabiosa_v.0_genomic.fna.gz

mv GCA_019182485.1_Rabiosa_v.0_genomic.fna.gz GCA_019182485.1_Rabiosa_v.0_genomic.fasta

Run BUSCO (Benchmarking Universal Single-Copy Orthologs) analysis on the Lolium multiflorum genome using the poales_odb10 database. BUSCO assesses the completeness of a genome assembly by searching for the presence of single-copy orthologous genes that are expected to be present in a specific database. The command is submitted to a job scheduler using the sbatch command and the script "busco_poales.sh".

sbatch ~/scratch/private/scripts/busco_poales.sh GCA_019182485.1_Rabiosa_v.0_genomic.fasta

Open a text editor to edit a file named "sorghum_gene.id". This file contains a list of 99 gene symbols for Sorghum bicolor identified from Ananda et al. 2021.

nano sorghum_gene.id

Use Entrez Direct to download the protein sequences corresponding to the 99 gene symbols in the "sorghum_gene.id" file. The command reads each line of the file using a while loop and creates a separate file for each gene symbol using the "touch" command. Then, for each gene symbol file, it uses the esearch command to search the NCBI protein database for the corresponding protein sequence and the efetch command to download it in FASTA format. The resulting protein sequence files are saved with the extension ".faa".

cat sorghum_gene.id | while read line; do touch $line; done

for file in *;
    do esearch -db protein -query $file | efetch -format fasta > "$file".faa
    done

Select the protein sequences that have only one isoform (i.e., a single copy of the gene) among the downloaded Sorghum bicolor protein sequences. The command searches each protein sequence file in the directory "sorghum/" for the number of FASTA headers (i.e., the number of protein isoforms) using the grep and wc commands. If there is only one header, the file is moved to a new directory named "1orth/".

for file in sorghum/*; do orth=$(grep '>' $file | wc -l); if [ $orth -eq 1 ]; then mv $file 1orth/; fi; done

Concatenate the protein sequences from the BUSCO analysis of Lolium multiflorum from both multi-copy and single-copy genes into a single file named "lolium_busco_prot.faa". The protein sequences are taken from two separate directories named "multi_copy_busco_sequences/" and "single_copy_busco_sequences/". Finally, the makeblastdb command is used to generate a blast database in protein format from the concatenated protein sequences.

for file in multi_copy_busco_sequences/*; do cat $file >> lolium_busco_prot.faa; done
for file in single_copy_busco_sequences/*; do cat $file >> lolium_busco_prot.faa; done

makeblastdb -in lolium_busco_prot.faa -dbtype prot -out lolium_prot

Perform a BLAST (Basic Local Alignment Search Tool) analysis of the Sorghum bicolor protein sequences against the Lolium multiflorum protein database created in the previous step. The blastp command is used to search the database with each protein sequence file in the "download/sorghum_prot/" directory using the options -max_target_seqs 3 to return up to three best hits, -outfmt 6 to output results in tabular format, -evalue 1e-3 to set the E-value threshold to 0.001, and -num_threads 2 to use two threads. The results are saved in a directory named "blast_results/" with a filename based on the gene symbol used to query the database.

for file in download/sorghum_prot/*;
    do fileshort=$(basename $file | sed s/".faa"//g)
    blastp -query $file -db lolium_prot -max_target_seqs 3 -outfmt 6 -evalue 1e-3 -num_threads 2 > blast_results/"$fileshort".outfmt6
    done

Move any BLAST output files that do not have any hits (i.e., zero-size files) to a directory named "no_hits/".

find . -size 0 -exec  mv {}  no_hits \;