why the CIRCexplorer2 is so slow

[z-bluesky]

When I ran CIRCxplorer2, it took me an over night to align the paired sequencing data using tophat. And then it took nearly 10 hours to go to the assembly step, and it hasn't finished running yet

[z-bluesky]

I spent 13 hours and 43 minutes in the assembly step, which is really time-consuming indeed, although I set the number of threads to 10000

[kepbod]

I recommend using STAR to align circRNAs and skipping the assembly step to speed up your process.

[z-bluesky]

I recommend using STAR to align circRNAs and skipping the assembly step to speed up your process.

Hi kepbod, I know that way can be faster, but I really need the tophat-align way to assemble and denovo to get novel RNA

[kepbod]

You can use STAR with stringtie to assemble and denovo to get novel circRNAs. You just neet to offer CIRCexplorer2 with the gene annotation file assembled using stringtie.

[spazm5]

You can use STAR with stringtie to assemble and denovo to get novel circRNAs. You just neet to offer CIRCexplorer2 with the gene annotation file assembled using stringtie.

Can you expand on this please? Because I did use a stringtie_assembly.gtf and it still only gave me known circRNAs and didn't work using denovo command.