Transcript mappability

[priyanka-surana]

Description of feature

This is ideally done with RNA-seq data. It is equal to percentage of reads mapped to the genome.
Possible software: WGSIM

[tkchafin]

According to this paper "mappability' = "the fraction of reads derived from a transcript that aligned to the original transcript" so I think this just involves taking the FASTA + GFF and extracting out the transcripts (as FASTA)

Some various tools I found so far:

• GenMap: https://academic.oup.com/bioinformatics/article/36/12/3687/5815974 ; has a module
• GEM mappability: https://evodify.com/gem-mappability/ ; has a module
• Umap (I think the one shown on UCSC mappability track): https://academic.oup.com/nar/article/46/20/e120/5086676
• mapCounter: https://github.com/shahcompbio/hmmcopy_utils has a module and uses bigwig

Could be worth doing some benchmarking to whittle out the 'slow' options? If the tools allows you to supply the transcript boundaries as bed or similar that might be a nice feature rather than extracting the transcriptome beforehand?