convert genomic bigWig file to transcriptome space

Question

[joe]

Hi all,

Is anyone aware of a function to convert a bw file mapped to a genome to map to a transcriptome (of said genome), where the input would be the genomic bw file and gff/gtf/bed annotation and output a single 'transcriptomic' bw file?

Thanks, Joe

[benformatics]

What exactly is a

single 'transcriptomic' bw

What exactly do you expect this file to contain? And what coordinate system do you think it will use?

[joe]

For example, the bw file would map to a multi-fasta file of the transcriptome, where each sequence is a transcript and each feature in the associated annotation file would start at 1 (or 0) and go to the end of the transcript. For my specific purpose, I need to align RNAseq data to a genome but I want to view the processed data in 'transcriptome space'.

[benformatics]

So a bigWig file is essentially coverages (numeric values) at specific genomic positions... Converting that to transcriptome space would be very difficult... also if you are hoping to differentiate between different isoforms that is going to be near impossible (or at bare minimum very hacky). You can convert genomic space to transcriptome space in R (https://rdrr.io/github/Bioconductor/GenomicFeatures/man/coordinate-mapping-methods.html) and might even be able to push/pull the coverage "meta data"... As a heads up I don't think it will be easy.

[joe]

Thanks for the tip! In my favour I mostly work with prokaryotes/archaea so the isoform issue is not so relevant.