Stringtie and prepDE

Question

[panacotaforcota]

Hello!

I try to use Stringtie to generate read coverage tables for prepDE.py. So, I do

./stringtie-2.0.3/stringtie file_sorted.bam -B -e -G merged.gtf > file.gtf

Is it correct that this command return gtf file or not?

Then I do

./prepDE.py -i file.gtf -g ./gene.counts.csv -t ./transc.counts.csv

and get it

Error: Text file with sample ID and path invalid (19 StringTie transcript 3033990 3034589 1000 . . gene_id "MSTRG.1"; transcript_id "MSTRG.1.1"; cov "3.420559"; FPKM "69.257027"; TPM "153.416428";)

Why?

[michael.ante]

I don't know by heart what stringtie writes to the stdout. I'd rather use the output option for the GTF than writing all output to a file.

The error message shows a line without strand information (+ or -). I'd check where that got lost.

[ATpoint]

Not helping with this problem but do you really need stringtie? If you work with well-annotated transcriptomes there is rarely a reason to assemble the transcriptome. THere are other tools to conveniently get a count matrix for DEG like featureCounts or quantifiers such as salmon or kallisto that you might want to check out.

Also, please use the code option to hightlight code and error messages. I did it for you this time.