###Purpose: Design guide RNAs for selected exons.
####Author: Lauren Sanders ####Version 1: 12/9/16
This program uses Max Haussler's guide RNA custom track on the UCSC Genome Browser to design guide RNAs for CRISPR screens targeting exons.
The program produces guide RNAs targeting three positions on each exon: mid-exon, 5' splice site, 3' splice site.

In order to provide flexibility regarding guide RNA quality score and guide RNA position, the 5' and 3' splice sites will have three gRNAs associated with them:

• gRNA with cutsite closest to splice site
  
• gRNA with highest score within 200 bp of splice site
  
• gRNA with next highest score within 200 bp of splice site
  

The mid-exon site will only have one gRNA, the one closest to the mid-exon point.

###INPUT
The program is designed to take as input a file containing only the Associated_Exon_Coordinates column of a JuncBase file, which is formatted like this: 1:1000-1100 (where 1=chromosome, 1000=exonStart, 1100=exonEnd).
Additionally, the program requires files containing CRISPR guide RNA sequences, downloaded from the Genome Browser. Part 1 of USAGE has instructions for getting these files.

###NOTE
If the JuncBase exon coordinates are different from the exon coordinates in the Genome Browser, the program may not be able to design gRNAs for that exon. In this case, a list of exons for which gRNAs were not designed will be output at the end of the program.

###USAGE
####Part 1: Crispr Files (ONE-TIME USE) a. Download the bigBed file crispr.bb from http://hgdownload.cse.ucsc.edu/gbdb/hg19/crispr/

b. In order to download the bigBedToBed conversion tool, convert the bigBed file to a Bed File, and remove all gRNAs whose sequence is not unique in the genome, perform the following Unix commands:

wget http://hgdownload.cse.ucsc.edu/admin/exe/linux.x86_64/bigBedToBed

chmod +x bigBedToBed

./bigBedToBed crispr.bb crispr.bed

sed '/MIT Spec. Score: -1/d' ./crispr.bed | sed '/Sequence is not unique in genome/d' | cut -f 1,2,3,6,12,14,15,16 >crispr_parsed.bed

c. run this command (runtime ~ 3 hrs) to parse the crispr file into individual chromosome files:

python3 parse_crispr.py <crispr_parsed.bed

####Part 2: gRNA Design (Every time you need to design gRNAs for a new exon set)

a. Retrieve only the Associated_Exon_Coordinates column from your JuncBase file, and write it into a new file.

b. Remove the header using this command:

sed -i '1d' infile

c. Use this command to run the gRNA design script (runtime 1-3 hours depending on number of exons. Running with nohup recommended):

python3 guideRNAselection.py -f infile

###OUTPUT (3 files)

1. infile_5PrimeGuideRNAs.csv
   
2. infile_3PrimeGuideRNAs.csv
   
3. infile_MidExonGuideRNAs.csv
   

The 5' and 3' files each have 12 columns as follows:

1. chromosome = exon chromosome number
   
2. exonStart = exon start coordinate
   
3. exonEnd = exon end coordinate
   
4. nearestGuide = guide nearest exon splice site regardless of score
   
5. nearestMIT = MIT score for the nearestGuide
   
6. nearestDoench = Doench score (percentage) for the nearestGuide
   
7. greenGuide = guide nearest splice site with Doench>60 and MIT>50
   
8. greenMIT = MIT score for greenGuide
   
9. greenDoench = Doench score (percentage) for the greenGuide
   
10. yellowGuide = guide nearest splice site with 30<Doench<60 and MIT>50
    
11. yellowMIT = MIT score for yellowGuide
    
12. yellowDoench = Doench score (percentage) for the yellowGuide
    

The MidExon file has 6 columns: chromosome, exonStart, exonEnd, MidExonGuide, guideMIT and guideDoench.