Subset of genes with at least 1 intron larger than 5000 kbp
1
2
Entering edit mode
3.3 years ago
shania90.lk ▴ 30

Hi to the experts,

Can I please get your help in trying to extract out genes that have at least 1 intron larger than 5 kbp? I have the GTF file and the reference genome for this species. What I want to know is, how many genes there are in the genome with at least 1 intron that has a length > 5 kbp. I have a feeling that GRanges object from the GTF file could be of use but still was unable to figure it out. I'm not sure whether I am being clear enough with my question though. Please feel free to ask questions to kindly help me find an answer for mine...

Thanks heaps, Shani.

GTF intron_length R Bash genomicFeatures • 2.1k views
2
Entering edit mode

You should have a look at this post on how to extract introns from a GTF file. From there on, you can simply use something like awk to filter out every intron longer than a certain length, and then extract the gene names from it. There is no need for R, everything can be done from the Unix command line. Try to play around a bit, and feel free to come back in case of questions.

1
Entering edit mode

AFAIK, largest intron in human is less than 1 mb. You were asking about 5mb (5000 kb) intron. What is the organism?

1
Entering edit mode

I'm betting on a typo (== 5000bp) ... ?

0
Entering edit mode

Yes, you are right.... 5 kbp it is... thanks for pointing that out... this is for the barley genome though...

9
Entering edit mode
3.3 years ago
benformatics ★ 2.6k

This can be done in R with the library GenomicFeatures.

library(GenomicFeatures)

## make TxDb from GTF file
txdb <- makeTxDbFromGFF('Homo_sapiens.GRCh38.93.gtf')

## get intron information
all.introns <- intronicParts(txdb)

## subset for introns greater than 5 kb (side-note: there are no introns longer than 5000 kbp in the human hg38 genome)
subset.introns <- all.introns[width(all.introns) > 5000]

## look at the number of unique gene names present
##(what i do is unlist some metadata... convert it to strings and then count the unique gene_name occurrences) I agree its a bit hard to read...

length(unique(as.character(unlist(mcols(subset.introns)\$gene_id))))
19469


In summary there are 61660 annotated introns > 5000 bp in size. These are derived from 19469 genes.