Qiime dada2 denoise-paired takes too long

Hello guys
I am analyzing illumina paired-end seq data with dada2.
It’s been running [qiime dada2 denoise-paired] command for over 24hrs.
I think it takes too long and I wonder if it’s going well…

1 Like

Good morning Jeongsu,

That first step can take a long time, and can use a lot of system memory (RAM). You can use Activity Monitor to check if this command is still running, or if it has crashed.

Let me know what you find! Specifically, we are looking for current CPU usage, RAM usage, and swap usage. If all is going well then CPU usage will be high, RAM usage will be medium, and swap usage will be near zero.

Colin

1 Like

Hi @Jeongsu_Kim - @benjjneb has some rough benchmarks for expected runtime on the DADA2 docs, I would recommend taking a look there for some ideas. As well, in the future you can add the --verbose flag while executing, this will show you a live log of the output, while it is running. Lastly, check out the --p-n-threads flag — changing the default value here will increase the number of threads DADA2 is executed with, which can reduce the overall runtime. Keep us posted! :t_rex:

3 Likes

As you suggested I added --verbose and it was very helpful that I could see if it’s running well.
And I got the results. It took 4 days though…

Thanks!

1 Like

Hi guys!

I am actually having the same issue with 143 paired-end samples, still running since Friday :open_mouth: (screenshots below from the last 7 days and today). When I run it I forgot to add --verbose and I also added the next scripts (alignment and phylogeny), so I really hope that fact is not messing up DADA2 step... and it finish soon! Anyway, here the usages, please let me know if they don't look fine. Many thanks!

Ana.

Hi again,

DADA2 step (+alignment and phylogeny) finished yesterday at 6 pm with success after 4 days analyzing 143 paired-end samples.

1 Like

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.