Base editor screening for gene functions

Summary

Base editor-based gene disruption tend to have less noise than cas9-based methods. Below is the design flowchart.

../../_images/BE_gene_design.png

Input

The program needs the following inputs:

  • Gene Name. Type: str.

  • PAM. Type: str.

  • ATG positions. This can be a dictionary for position annotation. For example, GATA1 has two start codons, one for GATA1 full length, the other for GATA1 short. Type: dict.

  • iSTOP prediction results. Type: file

Output

Output file name is {{gene_name}}.gRNA.annot.tsv

Candidate gRNA bed6 file, with additional columns of is_iSTOP and is_ATG. iSTOP is a tool for STOP codon mutation gRNA design. is_ATG indicates if this gRNA disrupts the start codon.

Usage

BE_gRNA_design.py GATA1 NG "{'GATA1Full': [48791110, 48791111], 'GATA1s': [48791873, 48791874]}" ../GATA1-istop.csv


BE_gRNA_design.py RPS19 NG "{'RPS19_main': [41860775, 41860776], 'RPS19_ENST00000221975': [41869081, 41869082]}" /home/yli11/Share/John/RPS19-istop.csv

R code to run iSTOP

CHANGE gene name as needed.

../../_images/iSTOP_design.png 
library(tidyverse)
library(iSTOP)
CDS_Human <- iSTOP::CDS_Hsapiens_UCSC_hg38()
Genome    <- BSgenome.Hsapiens.UCSC.hg38::Hsapiens
BE_spacing=PAM_spacing(optimal = c(14, 14), good = c(13, 15), ok = c(12, 17))
PAM_list=PAM_patterns_default(
  sgNGG = PAM_pattern(".GG", 3),
  sgNG = PAM_pattern(".G", 2),
  sgNGA = PAM_pattern(".GA", 3),
  sgNGCG = PAM_pattern(".GCG", 4),
  sgNGAG = PAM_pattern(".GAG", 4),
  sgNNGRRT = PAM_pattern("..G[AG][AG]T", 6),
  sgNNNRRT = PAM_pattern("...[AG][AG]T", 6),
)
GATA1 <-
  CDS_Human %>%
  filter(gene == 'GATA1') %>%
  locate_codons(Genome) %>%
  locate_PAM(Genome,PAM=PAM_list,spacing=BE_spacing)
readr::write_csv(GATA1,"GATA1-istop.csv")

code @ github.