Is there any tool/API available to convert GTF/GFF to JSON format?

Question

[lakhujanivijay]

Hi all

Before I start writing my own code to convert GTF file into a JSON format file, has anyone came across any API or tool to either convert or download the file is JSON ?

I am looking for a format like this - gene -> transcript -> exon :

"MOS": {
    "NM_005372.1": [
        {
            "exon_number": "1",
            "start": 57025501,
            "end": 57026541
        },
        {
            "exon_number": "1",
            "start": 57025504,
            "end": 57026541

Answer 1

[davmlaw]

Hi, I have written a Python library called PyReference which does exactly that!

It comes with a command pyreference_gff_to_json.py to turn GTF or GFF (RefSeq and Ensembl) into a gzipped JSON file.

There's also a Python wrapper around the JSON, which allows you to write genomics code more naturally.

Python isn't known for being fast but the library function for reading a JSON file is highly optimised - the following takes less than 4 seconds on my laptop:

import numpy as np
import pyreference

reference = pyreference.Reference()

my_gene_ids = ["MSN", "GATA2", "ZEB1"]
for gene in reference[my_gene_ids]:
    average_length = np.mean([t.length for t in gene.transcripts])
    print("%s average length = %.2f" % (gene, average_length))
    print(gene.iv)
    for transcript in gene.transcripts:
        if transcript.is_coding:
            threep_utr = transcript.get_3putr_sequence()
            print("%s end of 3putr: %s" % (transcript.get_id(), threep_utr[-20:]))

Answer 2

[Pierre Lindenbaum]

I wrote a gtf2xml http://lindenb.github.io/jvarkit/Gtf2Xml.html

Xml can be converted to json with a xslt stylesheet and xsltproc( see https://github.com/lindenb/xslt-sandbox/blob/master/stylesheets/bio/ncbi/pubmed2json.xsl for an example).

Answer 3

[Alex Reynolds]

Perhaps this answer might be of general use:

"Is there a JSON-based genomic feature format?" https://bioinformatics.stackexchange.com/questions/10386/is-there-a-json-based-genomic-feature-format/10387#10387

Using an existing format with a stable schema may be a better approach, especially if you will share these files with others.

Answer 4

[vkkodali_ncbi]

NCBI Datasets produces data report in json format that may contain all of the information you seek. You can download the command line tool and try it out as follows:

datasets download gene gene-id 5768 --filename test.zip

One of the output files ncbi_datasets/data/data_report.jsonl has the following:

"transcripts": [
    {
    "accessionVersion": "NM_002826.5",
    "cds": {
        "accessionVersion": "NM_002826.5",
        "range": [
        {
            "begin": "40",
            "end": "2283"
        }
        ]
    },
    "ensemblTranscript": "ENST00000367602.8",
    "exons": {
        "accessionVersion": "NC_000001.11",
        "range": [
        {
            "begin": "180154869",
            "end": "180155172",
            "order": 1
        },
        {
            "begin": "180166491",
            "end": "180166591",
            "order": 2
        },

Not quite a tool to convert an existing GTF/GFF3 to json format, but if you are dealing with NCBI Gene annotation, this can be an option.