#!/bin/bash
# 2021-01

#SBATCH --nodes=1			
#SBATCH --ntasks-per-node=1
#SBATCH --mem=64GB
#SBATCH --job-name=Make_nr_database


# Choose between long or short: A short job will start quickly after submitting, but will be terminated after 30 minutes

##SBATCH --time=24:00:00
#SBATCH -p short

export PERL5LIB=/data/pg-molgen/software/molgentools_mirror/lib
module load BioPerl/1.7.7-GCCcore-9.3.0


molgentools=/data/pg-molgen/software/molgentools_mirror

DIR_RAST=/data/p127804/PROJECTS/JanMaarten/Elisa/RAST_Annotation
DIR_small_orfs=/data/p127804/PROJECTS/JanMaarten/Elisa/Hybrid_Assembly_Stau_fasta
DIR_results=/data/p127804/PROJECTS/JanMaarten/Elisa/nr_db_Elisa_202101
DIR_blast_small_orfs=/data/p127804/PROJECTS/JanMaarten/Elisa/blast_small_orfs

mkdir $DIR_results


# 1 Merge all genomes
cat $DIR_RAST/*.faa > $DIR_RAST/all_RAST.fasta

# 2. replace | by _  to prevent errors in downstream processing
sed -i 's/|/\_/g' $DIR_RAST/all_RAST.fasta

# 3 RENAME RAST headers to names from the "lookup_table.txt"
$molgentools/genomics/g2d_conversions.pl -s $DIR_RAST -l all_RAST.fasta -r Cvrt_fasta2tab -o all_RAST.tab	
$molgentools/genomics/g2d_align_tables.pl -s $DIR_RAST -t $DIR_RAST/all_RAST.tab -k $DIR_RAST/lookup_table.txt -r Find_Replace -o all.tab
$molgentools/genomics/g2d_conversions.pl -s $DIR_RAST -l all.tab -r Cvrt_tab2fasta -o all.fasta	

# 4. Merge all small orfs that are not in the RAST annotation ==> 22.800 proteins
cat $DIR_blast_small_orfs/*.imperfect.match.faa > $DIR_blast_small_orfs/all.imperfect.match.fasta

# 5. Extend the all RAST proteins file with new small orfs
cat $DIR_blast_small_orfs/all.imperfect.match.fasta $DIR_RAST/all.fasta > $DIR_results/all_extended.fasta

# 6. remove redundant sequences
$molgentools/annotation/make_nr_db.pl -i $DIR_results/all_extended.fasta -o $DIR_results/all_nr.fasta
	# Entries in Query: 38388
	# Entries in Output: 10771




# Annotate proteins 
# NOTE; run 01.blast_uniprot.sh