7_Filtering.md

7 - Filtering

7.0 - Setup

7.0.1 - Define variables

genome genome fasta sequence.

n_jobs number of jobs to run in parallel.

7.0.2 - Set WD

cd 2-Annotation/2_2-Prediction/2_2_8-Filtering/

7.0.3 - Set files

mkdir fasta
mkdir blast

7.1 - Filter

Generate commands.

awk 'BEGIN {getline; id=$1 ;gsub(">","",id); print id ;filename="fasta/"id".fasta"; print $0 >filename } { if ($1~"^>") {id=$1 ; sub(/>/,"",id); filename="fasta/"id".fasta"; print id};  print $0 > filename }' gene_models.protein.fasta | sed 's:\(.*\):\blastp -num_threads 6 -db /DATA/db/RefSeq_protein_Plants/RefSeq_Plants.protein.faa -show_gis -outfmt 5 -query fasta/\1.fasta > blast/\1.xml:' > blast_commands.sh

Run blast commands in parallel.

parallel -j $n_jobs :::: blast_commands.sh > log

Run Blast2GO to load xml files and detect models without hits.

Generate the table No_blast.txt.

Filter out unwanted models and correct GFF3 file.

grep -vFf <(less to_remove.txt | cut -f 2 | sed '1,1d' ) gene_models.gff3  > gene_models.filtered.gff3

python /Scripts/GFF_extract_features.py -a gene_models.filtered.gff3 -g $genome -p gene_models.filtered -lscin > gene_models.filtered.log 2> gene_models.filtered.err