Pipeline for identification for novel human Genomic Safe Harbor (GSH) sites. The following criteria were used to computationally predict novel GSH:
- 50kb away from known genes
- 300kb away from known oncogenes
- 300kb away from microRNAs, centromeres, telomeres, genomic gaps
- 150kb away from lncRNAs, tRNAs
- 20kb away from enhancers
Produced genomic coordinates serve as computationally predicted GSH sites based on previously established as well as newly introduced criteria. All criteria used are universal for all cell types.
- wig2bed and gtf2bed from BEDOPS https://bedops.readthedocs.io/en/latest/content/installation.html#installation
- bedtools https://bedtools.readthedocs.io/en/latest/content/installation.html
- GENCODE annotation of the human genome version 24: comprehensive gene annotation, long non-coding RNA gene annotation, and predicted tRNA genes https://www.gencodegenes.org/human/
- Cancer Gene Census (CGC) genes from tier 1 (extensive evidence of association with cancer available) and tier 2 (strong indications of the association exist) https://cancer.sanger.ac.uk/census
- microRNAs from MirGeneDB (based on Fromm et al, A Uniform System For The Annotation Of Human microRNA Genes And The Evolution Of The Human microRNAome. Annu Rev Genet. 2015, 49: 213–242) https://mirgenedb.org
- Enhancers were obtained from EnhacerAtlas 2.0 http://www.enhanceratlas.org/downloadv2.php. Coordinates for human enhancers are given for the hg19 genome assembly. To convert to the GRCh38/hg38 version, use the LiftOver tool from the UCSC genome browser https://genome.ucsc.edu/cgi-bin/hgLiftOver
- Centromeric regions http://hgdownload.cse.ucsc.edu/goldenPath/hg38/database/centromeres.txt.gz
- Gaps in the genome sequence which include telomeres http://hgdownload.cse.ucsc.edu/goldenPath/hg38/database/gap.txt.gz
- Genome sequence, GRCh38.p5 assembly https://www.ncbi.nlm.nih.gov/projects/r_gencoll/ftp_service/nph-gc-ftp-service.cgi/?HistoryId=MCID_5fe9f269a15a7a0665583144&QueryKey=1&ReleaseType=RefSeq&FileType=GENOME_FASTA&Flat=true
Usage:
./predict_gsh.sh [-genes] [-oncogenes] [-micrornas] [-trnas] [-lncrnas] [-enhancers] [-centromeres] [-gaps] [-dist_from_genes] [-dist_from_oncogenes] [-dist_from_micrornas] [-dist_from_trnas] [-dist_from_lncrnas] [-dist_from_enhancers] [-dist_from_centromeres] [-dist_from_gaps] [-h|--help] Options:
-genes: Whether to exclude regions with and around genes (default=true)
-oncogenes: Whether to exclude regions with and around oncogenes (default=true)
-micrornas: Whether to exclude regions with and around microRNAs (default=true)
-trnas: Whether to exclude regions with and around tRNAs (default=true)
-lncrnas: Whether to exclude regions with and around lncRNAs (default=true)
-centromeres: Whether to exclude regions with and around centromeres (default=true)
-gaps: Whether to exclude regions with and around gaps (default=true)
-enhancers: Whether to exclude enhancer regions (default=true)
-dist_from_genes: Minimal distance from any safe harbor to any gene in bp (default=50000)
-dist_from_oncogenes: Minimal distance from any safe harbor to any oncogene in bp (default=300000)
-dist_from_micrornas: Minimal distance from any safe harbor to any microRNA in bp (default=300000)
-dist_from_trnas: Minimal distance from any safe harbor to any tRNA in bp (default=150000)
-dist_from_lncrnas: Minimal distance from any safe harbor to any long-non-coding RNA in bp (default=150000)
-dist_from_enhancers: Minimal distance from any safe harbor to any enhancer in bp (default=20000)
-dist_from_centromeres: Minimal distance from any safe harbor to any centromere in bp (default=300000)
-dist_from_gaps: Minimal distance from any safe harbor to any gaps in bp (default=300000)
-h, --help: Prints helpRunning with the default parameters:
./predict_gsh.sh
Output:
Getting gene annotation from GENCODE
Distance from genes = 50000 bp
Distance from oncogenes = 300000 bp
Distance from microRNAs = 300000 bp
Distance from tRNAs = 150000 bp
Distance from lncRNAs = 150000 bp
Distance from enhancers = 20000 bp
Distance from centromeres = 300000 bp
Distance from gaps = 300000 bp
Merging all genomic regions to avoid
Obtaining genomic coordinates and sequences of safe harbors
Running with modified parameters:
./predict_gsh.sh -trnas false -dist_from_lncrnas 50000 -dist_from_enhancers 0 -dist_from_centromeres 0 -dist_from_gaps 0Output:
Getting gene annotation from GENCODE
Distance from genes = 50000 bp
Distance from oncogenes = 300000 bp
Distance from microRNAs = 300000 bp
Distance from lncRNAs = 50000 bp
Distance from enhancers = 0 bp
Distance from centromeres = 0 bp
Distance from gaps = 0 bp
Merging all genomic regions to avoid
Obtaining genomic coordinates and sequences of safe harbors
The output is two files: Safe_harbors.bed that has genomic coordinates of all regions potentially containing safe harbors and Safe_harbors.fasta contains sequences of these regions.
Aznauryan et al. (2022), Discovery and validation of novel human genomic safe harbor sites for gene and cell therapies. Cell Genomics