search

DavidsonGroup / flexiplex

The Flexible Demultiplexer
https://davidsongroup.github.io/flexiplex/
MIT License
23 stars 2 forks source link

Processing time #40

Closed nrhorner closed 6 months ago

nrhorner commented 6 months ago

Hi,

I'm trying to run some 10x reads though flexiplex, but it's taking quite a long time to process. 1,000 reads took 17mins. I'm using flexiplex v1.0.1 on Ubuntu 20.04

This is the commend used: flexiplex -d 10x3v3 -k 3M-february-2018.txt -p 32 1k_test.fq

3M-february-2018.txt is the 3Prime v3 whitelist from 10x

Maybe I'm doing something wrong?

Thanks,

Neil

nadiadavidson commented 6 months ago

Hi Neil,

It's taking so long because the whitelist from 10x is very large. If you don't know the cell barcodes (non-empty drops) from matched single cell data, you can discvoer these first from the data, using the instructions at: https://davidsongroup.github.io/flexiplex/#assigning-long-reads-to-10x-barcodes-when-barcodes-are-unknown This should be much fasters. Please let us know if you run into any other issues.

Cheers, Nadia.

nrhorner commented 6 months ago

Hi Nadia,

Thanks for the info. I will try the two phase solution as suggested.

Cheers,

Neil

nrhorner commented 6 months ago

That's more like it. Very fast indeed!

Cheers, Neil