hassanfa / GFFtoolkit

Extracts information from GFF and prepare them in bed format

GFFtoolkit

Extracts information from GFF and prepare them in bed format

Required tools

This tool requires the following from kentutils, which should be downloaded from http://hgdownload.soe.ucsc.edu/admin/exe/ or any other related source.

• gff3ToGenePred
• genePredToBed
• bedGeneParts
• tabQuery

For linux_x86 machines either downloaded all the tools or download them individually:

Note: assumes you have $HOME/bin or you want these tools to be in your $PATH. Otherwise baby, you should skip replace ${HOME}/bin/ in the script below with to whatever path you want.

for i in gff3ToGenePred genePredToBed bedGeneParts tabQuery
do
  rsync -aP \
     rsync://hgdownload.soe.ucsc.edu/genome/admin/exe/linux.x86_64/${i} ${HOME}/bin/
done

Running GFFtoolkit

The following script will create a bed file from a sample GFF3 file within data directory. The output structure is explained later on.

bash GFFtoolkit.sh -i data/gencode.v19.annotation.chr22.gff3 -o test/

Input data

• Input data is a GFF3 file downloaded from gencode. You can download your own GFF3 for hg19 using the command below:

wget ftp://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_19/gencode.v19.annotation.gff3.gz

Output data

From the files above, the most important ones are:

1. transcript_table: A table of transcripts found in the GFF3 files with extra column information. This file can be queried using tabQuery.
2. gene_table: A table of genes found in the GFF3 files with extra column information. This file can be queried using tabQuery.
3. gencode.v19.annotation.exons.extended: If the input file is named the gencode.v19.annotation.gff3 the GFF3\_rootname.exons.extended is list of exons for each transcript.
4. gencode.v19.annotation.exons.extended.simpleid: If the input file is named the gencode.v19.annotation.gff3 the GFF3\_rootname.exons.extended.simpleid is list of exons for each transcript but strips version number from these IDs.

Running the example above will generate a result folder within this repo. It within that path, it will generate the following files:

├── attrsOut
├── attrsOut_table
│   ├── attrsOut -> ../attrsOut
│   ├── gene
│   │   ├── gene_id
│   │   ├── gene_name
│   │   ├── gene_status
│   │   ├── gene_type
│   │   ├── havana_gene
│   │   ├── ID
│   │   ├── level
│   │   ├── tag
│   │   ├── transcript_id
│   │   ├── transcript_name
│   │   ├── transcript_status
│   │   └── transcript_type
│   ├── gene_table
│   ├── transcript
│   │   ├── ccdsid
│   │   ├── gene_id
│   │   ├── gene_name
│   │   ├── gene_status
│   │   ├── gene_type
│   │   ├── havana_gene
│   │   ├── havana_transcript
│   │   ├── ID
│   │   ├── level
│   │   ├── ont
│   │   ├── Parent
│   │   ├── protein_id
│   │   ├── tag
│   │   ├── transcript_exon_count
│   │   ├── transcript_id
│   │   ├── transcript_length
│   │   ├── transcript_name
│   │   ├── transcript_status
│   │   └── transcript_type
│   └── transcript_table
├── gencode.v19.annotation.chr22.bed
├── gencode.v19.annotation.chr22.exons
├── gencode.v19.annotation.chr22.exons.extended
├── gencode.v19.annotation.chr22.exons.extended.simpleid
└── gencode.v19.annotation.chr22.gp