CAD data analysis scripts

#!/bin/bash

task=$1

outdir=$2

ref_fasta=$3

python form_svm_input_fastas.py \

        --outf $outdir/$task/svm.inputs.$task.test.0 $outdir/$task/svm.inputs.$task.test.1 $outdir/$task/svm.inputs.$task.test.2 $outdir/$task/svm.inputs.$task.test.3 $outdir/$task/svm.inputs.$task.test.4 $outdir/$task/svm.inputs.$task.test.5 $outdir/$task/svm.inputs.$task.test.6 $outdir/$task/svm.inputs.$task.test.7 $outdir/$task/svm.inputs.$task.test.8 $outdir/$task/svm.inputs.$task.test.9 $outdir/$task/svm.inputs.$task.train.0 $outdir/$task/svm.inputs.$task.train.1 $outdir/$task/svm.inputs.$task.train.2 $outdir/$task/svm.inputs.$task.train.3 $outdir/$task/svm.inputs.$task.train.4 $outdir/$task/svm.inputs.$task.train.5 $outdir/$task/svm.inputs.$task.train.6 $outdir/$task/svm.inputs.$task.train.7 $outdir/$task/svm.inputs.$task.train.8 $outdir/$task/svm.inputs.$task.train.9 \

        --neg_pickle $outdir/$task/$task.candidate.negatives.gc.p \

        --overwrite_outf \

        --ref_fasta $ref_fasta \

        --peaks $outdir/$task/svm.peaks.$task.test.0.gc.seq $outdir/$task/svm.peaks.$task.test.1.gc.seq $outdir/$task/svm.peaks.$task.test.2.gc.seq $outdir/$task/svm.peaks.$task.test.3.gc.seq $outdir/$task/svm.peaks.$task.test.4.gc.seq $outdir/$task/svm.peaks.$task.test.5.gc.seq $outdir/$task/svm.peaks.$task.test.6.gc.seq $outdir/$task/svm.peaks.$task.test.7.gc.seq $outdir/$task/svm.peaks.$task.test.8.gc.seq $outdir/$task/svm.peaks.$task.test.9.gc.seq $outdir/$task/svm.peaks.$task.train.0.gc.seq $outdir/$task/svm.peaks.$task.train.1.gc.seq $outdir/$task/svm.peaks.$task.train.2.gc.seq $outdir/$task/svm.peaks.$task.train.3.gc.seq $outdir/$task/svm.peaks.$task.train.4.gc.seq $outdir/$task/svm.peaks.$task.train.5.gc.seq $outdir/$task/svm.peaks.$task.train.6.gc.seq $outdir/$task/svm.peaks.$task.train.7.gc.seq $outdir/$task/svm.peaks.$task.train.8.gc.seq $outdir/$task/svm.peaks.$task.train.9.gc.seq

#!/bin/bash

#get the inverse intersection of all peaks and all gc genome bins

task=$1

peaks=$2

outdir=$3

genomewide_gc=$4

rm -f $outdir/$task/$task.all.positives.bed

for split in `seq 0 9`

do

    cat $outdir/$task/svm.peaks.$task.test.$split.gc.seq | cut -f 1-3 >> $outdir/$task/$task.all.positives.bed

done

python peak_to_bed.py $peaks $outdir/$task/$task.peaks.bed

#cut -f 1-3 $peaks > $outdir/$task/$task.peaks.bed

cat $outdir/$task/$task.peaks.bed $outdir/$task/$task.all.positives.bed > $outdir/$task/$task.all.positives.peaks.bed

bedtools intersect -v -a $genomewide_gc -b $outdir/$task/$task.all.positives.peaks.bed > $outdir/$task/$task.candidate.negatives.tsv

#!/bin/bash

task=$1

outdir=$2

python get_chrom_gc_region_dict.py \

       --input_bed $outdir/$task/$task.candidate.negatives.tsv \

       --outf $outdir/$task/$task.candidate.negatives.gc.p