This code designs every possible guide in the CDS region, 30 nucleotides into the intronic and UTR regions for user-defined transcripts. It then annotates the possible edits for each guide. Separate files annotating ClinVar SNPs are also generated.
Author: Mudra Hegde, Ruth HannaEmail: mhegde@broadinstitute.org, rhanna@broadinstitute.org
Version: 2.0
Inputs
- Input File:.txt file with list of Ensembl transcript IDs in the first column and gene symbols in the second column OR FASTA file with nucleotide sequence
- Variant File: Variant file: File with ClinVar SNPs (variant_summary.txt). This file can be downloaded from ftp://ftp.ncbi.nlm.nih.gov/pub/clinvar/tab_delimited/variant_summary.txt.gz.
- Input type: Indicate whether the file contains a list of transcripts or a nucleotide sequence.
- Base editor type: Indicate the type of base editor (Rees et al.,2018) for which designs are required. This choice dictates the choice of PAM, sgRNA length, editing window and type of edit.
- PAM: PAM preference if BE type has not been selected; Default: NGG.
- Edit window: Editing window relative to nucleotide position in sgRNA, if BE type has not been selected; Default: 4-8.
- sgRNA length: Length of sgRNA excluding PAM sequence, if BE type has not been selected; Default:20.
- Edit: Type of edit made by base editor, if BE type has not been selected; Default: all, annotates for both C->T and A->G edit.
- Intron buffer: Number of bp into the intron to consider for guide design.
- Filter GC: Whether to filter out edits in a GC motif.
- Output name: Name for output folder.