JACKIE2

JACKIE (Jackie and Albert's CRISPR K-mer Instances Enumerator) [yes, a recursive acronym!], is a software pipeline written mainly in C++ with accessory scripts written in bash and python languges, that allows enumeration of all potential SpCas9 binding sites in a genome and output their sequences, copy numbers and locations. We have demonstrated its application to design sgRNA with clustered repetitive binding sites for imaging genomic region. Please see the preprint for more details.

Precomputed CRISPR sites (JACKIEdb)

Precomputed CRISPR sites (JACKIEdb) available for hg38 and mm10, available at http://crispr.software/JACKIE Users interested in strating from precomputed sites to further filter or identify sites within regions, etc, can jump to the later sections

Installation

With root privilege:

./configure
make
make install

Install to specific path:

./configure --prefix=/path/to/install/
make
make install

Add JACKIE to path. In your ~/.bashrc file, add a line:

export PATH=/path/to/install/:${PATH}

Example run

Download genome fasta files and produce a merged files for "non-random" chromosomes

genome=<fill in your genome> #e.g., hg38
genomesRoot=<fill in your genomes data root path> 
pathToGenome=$genomesRoot/$genome
genomeFasta=$pathToGenome/$genome.nr.fa

cd $pathToGenome
wget --timestamping "ftp://hgdownload.cse.ucsc.edu/goldenPath/$genome/chromosomes/*"
gunzip *.gz
mkdir random
mv *random*.fa random/
mv chrUn*.fa random/
mkdir nr
mv *.fa nr
cat nr/*.fa > $genome.nr.fa

Run first step of JACKIE, assuming your cluster uses qsub:

genome=<fill in your genome> #e.g., hg38
genomesRoot=<fill in your genomes data root path> 
pathToGenome=$genomesRoot/$genome
genomeFasta=$pathToGenome/$genome.nr.fa

jackieDB=$pathToGenome/jackieDB/
mkdir $jackieDB

#generate binary represetation of sgRNA binding locations

for N in A C G T; do
echo "date; JACKIE -b2 $genomeFasta $jackieDB .bin 6 $jackieDB/$N.ref.txt $N n; date" | qsub -l walltime=48:00:00
done

Please make sure all four jobs for step one have completed successfully, e.g., by using qstat and checking output files. Second step, load binary files (*.bin), sort by sequence, output bed files:

#output bed file from binary files.
for prefix in AA AC AT AG CA CC CT CG TA TC TT TG GA GC GT GG; do
echo "date; outbedForPrefixJob.sh $jackieDB $prefix 1 0; date" | qsub -l walltime=24:00:00 -e `pwd`/$prefix.stderr.txt -o `pwd`/$prefix.stdout.txt	
done

Please make sure all jobs in second step have completed successfully. Third step, merge all bed files into one:

#concatenate all bed files into one
echo "cat $jackieDB/*.bed > $jackieDB/${genome}PAM.BED" | qsub -l walltime=24:00:00

Optional step - collapse CRISPR sites from same chromosome into an extended bed format, with each line containing sites of the same sequence:

#collapse sgRNA binding locations with same sequecnes into an extended bed format
echo "chainExonBedsToTranscriptBed.py $jackieDB/${genome}PAM.BED 0 > $jackieDB/${genome}PAM.sameChr.tx.bed" | qsub -l walltime=24:00:00

Optional step - sort extended bed file:

#sort extended bed file
echo "sort -k1,1 -k2,2n $jackieDB/${genome}PAM.sameChr.tx.bed > $jackieDB/${genome}PAM.sameChr.tx.sorted.bed" | qsub -l walltime=24:00:00

#some CRISPR sites are overlapping, and will crash browser visualization. Remove entries with overlapping sgRNA sites
echo "removeIllegalBlockEntries.py $jackieDB/${genome}PAM.sameChr.tx.sorted.bed $jackieDB/${genome}PAM.sameChr.tx.sorted.legal.bed $jackieDB/${genome}PAM.sameChr.tx.sorted.illegal.bed" | qsub -l walltime=24:00:00

Filtering examples

Select clustered sgRNA with (minBS)5 to (maxBS)8 binding sites and within (minDist)5kb to (maxDist)10kb distance Precomputed hg38 same-chromosome sites and mm10 same-chromosome sites

#select clustered sgRNA with (minBS)5 to (maxBS)8 binding sites and within (minDist)5kb to (maxDist)10kb distance
minBS=5
maxBS=8
minDist=5000
maxDist=10000
genome=hg38 #human genome example
awk -v FS="\t" -v OFS="\t" -v minBS=$minBS -v maxBS=$maxBS -v minDist=$minDist -v maxDist=$maxDist '($3-$2>=minDist && $3-$2<=maxDist && $5>=minBS && $5<=maxBS)' $genome.sameChr.tx.sorted.legal.bed > $genome.sameChr.tx.sorted.legal.bed

Select unique sgRNA sites

#select unique sgRNA sites
awk -v FS="\t" -v OFS="\t" '($5==1)' $jackieDB/${genome}PAM.BED > $jackieDB/${genome}PAM.1copy.BED

Precomputed hg38 1-copy sites and mm10 1-copy sites

Selecting (clustered or single) CRISPR sites overlapping regions of interest

Put regions of interest into a bed file, say, selection.bed:

chr13	75845001	75850000	Region_A
chr13	76493001	76498000	Region_B

Overlap selection.bed with one copy sites:

fastjoinBedByOverlap.py selection.bed  $jackieDB/hg38PAM.1copy.BED > selection.overlap.hg38PAM.1copy.BED

Run Cas-OFFinder

Requires offline version of Cas-OFFinder at http://www.rgenome.net/cas-offinder/portable) Also, cas-offinder should be in $PATH

For example, from selection.overlap.hg38PAM.1copy.BED above: Let say, up to 3 mismatches. Sequence is encoded in the itemName (on the 8th column, second component of a split with "/"), so seqColExtract=8,/,2


genome=<fill in your genome> #e.g., hg38
genomesRoot=<fill in your genomes data root path> 
pathToGenome=$genomesRoot/$genome

BEDFile=selection.overlap.hg38PAM.1copy.BED
cas_outDir=/path/to/IOForCasOffFinder
maxNmismatches=3
seqColExtract=8,/,2
runCasOFFinderOnSequences.py $BEDFile $seqColExtract $maxNmismatches $pathToGenome $cas_outDir > $BEDFile.cas_off.txt

albertwcheng / JACKIE