Ben Johnson

Continuing this thread - would it be worth it if a few of us here put our heads together and attempted training a [direct] RNA model? I know there are...

@jennymoon90 the GTF you could use for hg19 would probably be from Gencode (https://www.gencodegenes.org/human/release_19.html). You can then import this into R with: ``` library(rtracklayer) library(GenomicRanges) gencode_hg19_gtf

Is this something that seems feasible @k3yavi or @rob-p?

Thanks so much @k3yavi! We will take a look and see about adding a small function to add a synthetic cell barcode in front of the UMI on read 2....

So the way around this @tk2 and @kenza12 is to download the latest version of exonerate (https://www.ebi.ac.uk/about/vertebrate-genomics/software/exonerate) [2.4.0], recompile and execute retroseq. Must be some issue with 2.2.0...

Very much looking forward to the bioconda recipe being available. Thanks again for writing this!