How to import a gtf file in R so that it appears in tabular format as if it were a data frame?

Question

[Alex]

Hi all!

How to import a gtf file in R so that it appears in tabular format as if it were a data frame?

This is the file name:

 GCF_023701775.1_HaSCD2_genomic.gtf

This is the file head:

#gtf-version 2.2

#!genome-build HaSCD2

#!genome-build-accession NCBI_Assembly:GCF_023701775.1

#!annotation-source NCBI RefSeq Helicoverpa armigera Annotation Release 101 

NC_064776.1 Gnomon  gene    4347    6338    .   -   .   gene_id "LOC126054536"; transcript_id ""; db_xref "GeneID:126054536"; description "uncharacterized LOC126054536"; gbkey "Gene"; gene "LOC126054536"; gene_biotype "protein_coding";

[seidel]

It's just a tab-separated file. You can use read.delim(filename, sep="\t", skip=n) where n = header lines. Is that really what you want?

[ATpoint]

Actually tab-separated with nested columns separated by semicolon, so the rtracklayer route should be preferred since it takes care of separating those columns-in-columns.

[seidel]

Thus my question to the OP..."Is that really what you want?" as it wasn't clear to me what they were trying to accomplish with a data frame. But yes, I figured the real solution would involve txdb or rtracklayer, which are both great for this.

Answer 1

[benformatics]

 library(rtracklayer)

 GTF <- rtracklayer::import('your.gtf')
 GTF <- data.frame(GTF)

although i would opt for a txdb if you are working in R

library(GenomicFeatures)
txdb <- makeTxDbFromGFF('your.gtf')

Answer 2

[acvill]

If you find that the functions from the rtracklayer or GenomicsFeatures libraries don't work for your annotations, you can try my read_gtf() function:

Transform a GTF file into a data frame in R