curation_2.0_pipe.sh

#!/bin/sh 

Help()
{
    # Display help

    printf "\nCommand: sh curation_2.0_pipe.sh -f <original fasta> -a <agp> <options>\n\n"
    echo "-h Prints help."
    echo "-f Pass original fasta file with combined haplotypes."
    echo "-a Pass the agp generated by PretextView."
    # echo "-p Pass the primary assembly you curated (1 for haplotype 1 (default), 2 for haplotype 2)."
    printf "\n"
}

fasta=""
agpfile=""
hap=""

while getopts ":hf:a:p:" option; do
    case $option in 
        h) #display Help
            Help
            exit;;
        f) #Pass original fasta file 
            fasta=$OPTARG;;
        a) #Pass Pretext generated AGP file of curated assembly
            agpfile=$OPTARG;;
    esac
done

if [ -d logs ]
then
    count=`ls logs/* | wc -l`
    exec 1<> logs/std.${count}.out
else 
    mkdir -p logs 
    exec 1<> logs/std.0.out
fi 

pth=$( cd -- "$( dirname -- "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )
printf "Rapid-curation-2.0 scripts located in: $pth\n"

## Programs/tools
# use_gfastats=/vggpfs/fs3/vgl/store/nbrajuka/gfastats/build/bin/gfastats
# use_seqkit=/vggpfs/fs3/vgl/store/nbrajuka/conda/envs/statistics/bin/seqkit
printf "Dependecies:\nBiopython v1.81\npandas\ngfastats v1.3.6\n" 

# could reasonably put everything in a function and just call for both haps. 

mkdir -p Hap_1
mkdir -p Hap_2

printf "\nOriginal assembly: ${fasta} \nPretextView generated AGP: ${agpfile}\n\n" ### but checks/breakpoints for if these aren't provided.

printf "Running AGPcorrect on the PretextView generated agp to correct for sequence lengths.\n" 
printf "python3 ${pth}/AGPcorrect.py ${fasta} ${agpfile}\n\n"
python3 $pth/AGPcorrect.py ${fasta} ${agpfile} 

printf "Splitting the haplotypes from the corrected AGP. Outputs sent to respective directories.\n"
printf "python3 $pth/hap_split.py\n\n"
python3 $pth/hap_split.py 

printf "Assigning unlocs before the agp is imposed on the fasta.\n"
## If the --agp-to-path in the next block is run first the unlocs will get assimilated into their main assigned scaffolds - they need to be differentiated first.
printf "python3 $pth/unloc.py Hap_1\n" 
printf "python3 $pth/unloc.py Hap_2\n\n"
python3 $pth/unloc.py Hap_1
python3 $pth/unloc.py Hap_2

printf "Imposing the haplotypic agp on the original fasta to generate a curated fasta.\n"
printf "gfastats $fasta --agp-to-path Hap_1/hap.unlocs.no_hapdups.agp --sort largest -o Hap_1/hap.sorted.fa\n"
gfastats $fasta --agp-to-path Hap_1/hap.unlocs.no_hapdups.agp -o Hap_1/hap.unlocs.no_hapdups.fa 2>> logs/std.${count}.out 

printf "gfastats $fasta --agp-to-path Hap_1/hap.unlocs.no_hapdups.agp --sort largest -o Hap_2/hap.sorted.fa\n\n"
gfastats $fasta --agp-to-path Hap_2/hap.unlocs.no_hapdups.agp -o Hap_2/hap.unlocs.no_hapdups.fa 2>> logs/std.${count}.out 

printf "gfastats Hap_1/hap.unloc.no_hapdups.fa --sort largest -o Hap_1/hap.sorted.fa"
gfastats Hap_1/hap.unlocs.no_hapdups.fa --sort largest -o Hap_1/hap.sorted.fa 2>> logs/std.${count}.out

printf "gfastats Hap_2/hap.unloc.no_hapdups.fa --sort largest -o Hap_2/hap.sorted.fa"
gfastats Hap_2/hap.unlocs.no_hapdups.fa --sort largest -o Hap_2/hap.sorted.fa 2>> logs/std.${count}.out 

printf "\nSubstituting scaffold for chromosome assignments.\n"
printf "python3 $pth/chromosome_assignment.py Hap_1\n"
python3 $pth/chromosome_assignment.py Hap_1

printf "python3 $pth/chromosome_assignment.py Hap_2\n\n"
python3 $pth/chromosome_assignment.py Hap_2 


exec 1>&-