ENCODE FCC CRISPR working group

scripts

For checking file correctness

check_guide_quant_format.py

input: guide_quant_format.txt (guideline for ENCODE standard) and ENCODE-formatted guide_quant files

NOTE: "guide_quant_format.txt" in this repository might not be up-to-date (last modified: 5/10/2021).

Purpose: Check whether input file has a proper format that matches with guide_quant_format guideline.

sample command line: python {format description file} {test file}

python check_guide_quant_format.py guide_quant_format.txt GATA_rep1_HS_ENCODE_guideQuant.bed

sample output

If successful:

Test 1 passed

Test 2 passed

Test 3 passed

Example failure:

Test 1 passed

Test 2 failed. In line 4695, 15'th element must be either targeting or negative_control

check_PAM.py

input: ENCODE-formatted guide_quant files, a reference genome fasta file. ref hg38 can be downloaded from: https://www.encodeproject.org/files/GRCh38_no_alt_analysis_set_GCA_000001405.15/

Purpose: If your file passes check_guide_quant_fomat.py, use this script to check whether your PAM coordinates are correctly extracted by checking NGG sequence.

sample command line: python {ifile: guide_quant} {reference fasta}

python check_PAM.py MYC_rep1_LS_ENCODE_guideQuant.bed GRCh38_no_alt_analysis_set_GCA_000001405.15.fasta

Sample result 1 :

NGG 50732.0

other 398.0

More than 80% of the PAMs are NGG. The coordinates are likely to be correct

Sample result 2:

NGG 5433.0

other 4689.0

Less than 80% of the PAMs are NGG. The coordinates are likely to be incorrect

For analysis (PAM coordinate -> log2FC) -- after passing the above two tests.

gRNA_to_log2FC.py

input: ENCODE-formatted guide_quant files

Purpose: compute log2FC in three different way: raw log2FC, log2FC z-transformed using negative control guides, log2FC z-tranformed using all guides

output: log2FC_summary files

sample command line (python {gRNA quant 1 (e.g. T0)} {gRNA quant 2 (e.g. T14)} {ofile prefix})

python gRNA_to_log2FC.py MYC_rep1_LS_ENCODE_guideQuant.bed screens/MYC_rep1_HS_ENCODE_guideQuant.bed Sabeti_HCRFlowFISH_MYC_R1

summary_to_bedgraph.py

input: log2FC_summary file

Purpose: select a type of log2FC in log2FC_summary, and use it to generate .bedgraph. Further, to ensure that the genome coordinates do not overlap (for compatibility with UCSC genome browser), output PAM coordinates will have width of 1 (coordinate of 'N'GG for (+)-strand and of CC'N' for (-)-strand).

output: .bedgraph

sample command line (format {log2FC summary} {col index for log2fc} {ofile name})

for raw log2fc

python summary_to_bedgraph.py Sabeti_HCRFlowFISH_MYC_R1.log2FC_summary 2 Sabeti_HCRFlowFISH_MYC_R1_log2FC_0.bedgraph

for ztransformed_by_neg_control

python summary_to_bedgraph.py Sabeti_HCRFlowFISH_MYC_R1.log2FC_summary 3 Sabeti_HCRFlowFISH_MYC_R1_log2FC_1.bedgraph

for ztransformed_by_all

python summary_to_bedgraph.py Sabeti_HCRFlowFISH_MYC_R1.log2FC_summary 4 Sabeti_HCRFlowFISH_MYC_R1_log2FC_2.bedgraph

for mean-normalized log2((X/mean(X)+1)/(Y/mean(Y)+1))

python summary_to_bedgraph.py Sabeti_HCRFlowFISH_MYC_R1.log2FC_summary 5 Sabeti_HCRFlowFISH_MYC_R1_log2FC_3.bedgraph

for sum-normalized log2((X/sum(X)+1)/(Y/sum(Y)+1))

python summary_to_bedgraph.py Sabeti_HCRFlowFISH_MYC_R1.log2FC_summary 6 Sabeti_HCRFlowFISH_MYC_R1_log2FC_4.bedgraph

sample file format

ENCODE standard guide_quantification file

For more detail, check "guide_quant_format.txt"

chr12 54300767 54300770 GATA1|chr12:54300748-54300767:+ 366 + chr12:54300748-54300767:+ chrX 48786590 48786591 + GATA1 ENSG00000102145 GGATTCCAGTGAGATCCGAG GGATTCCAGTGAGATCCGAG targeting NA

chr12 54300811 54300814 GATA1|chr12:54300792-54300811:+ 551 + chr12:54300792-54300811:+ chrX 48786590 48786591 + GATA1 ENSG00000102145 CTCCACCACAGGTGCCTGAA GCTCCACCACAGGTGCCTGAA targeting NA

Only the first three, fifth, and sixth columns are relevant for these scripts.

First three: PAM coordinate

Fifth: total number gRNA sequences that are sequenced in a cell population

Sixth: strand location of the PAM

.log2FC_summary

name PAM_ID raw_log2FC ztransformed_by_neg_control ztransformed_by_all_guides

GATA1|chr12:54300748-54300767:+ chr12:54300767-54300770:+ -0.6898171127857369 -1.3909202490331116 -1.1558450982161879

GATA1|chr12:54300792-54300811:+ chr12:54300811-54300814:+ 0.14274017211608214 -0.3609625301812802 -0.3592771390999452

.bedgraph

chrX 48476136 48476136 0.16272950003810832

chrX 48476137 48476137 0.14542752477380644

scripts for gRNA coordinate conversions

Some of the CRISPR screens used in CRISPR WG had coordinates in hg19 (Engreitz/Bassik/Shen lab), and we converted their hg19 coordinates to hg38 using bowtie1. Some python and shell scripts are uploaded here, which can be executed in this order: bed_to_fa_all.sh -> bowtie_all.sh -> bwt_to_consersion_all.sh -> convert_cv_using_coord_all.sh. But, some of these scripts need to be re-configured to be usable (e.g. directory names specified in .sh).