Pairtools for OmniC pipeline no longer working about pairtools

[Phlya]

What is the exact error you get for the header? parse takes the bwa mem output, nothing we can do about that...

128 Gb is a lot, is it possible your duplication rate is super high? Regardless, one of the options above should help with that.

[Phlya]

Can you clarify at what stage do you have to run pairtools generate header?

How much memory do you have available that you are getting the out of memory error? To avoid any out of memory errors, you can use the old backend for dedup, using --backend cython in the dedup command. Or, instead, you can reduce the chunksize, e.g. --chunksize 10000.

[gdaviduu]

Thanks for the quick reply!
An error cropped up at pairtools parse requiring me to run pairtools generate header, but I instead downgraded to the old version of pairtools v.1.0.1. and pinned the older version of numpy. This seems to have circumvented the error.

If I use a full normal node, I get 128 Gb of RAM.

[liphardt]

I encountered the same error during the dedup stage. The error comes from stats.py where np.int is deprecated and just int should be used when declaring the type. The error:

AttributeError: module 'numpy' has no attribute 'int'.
`np.int` was a deprecated alias for the builtin `int`. To avoid this error in existing code, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.

I also had the header error but it was a result of the dedup stage failing so the output of that, the input for the next step, was empty.

Hope this helps.

[Phlya]

From what I can tell, all instances of np.int have been eliminated from the master branch. In the next release this error won't appear.

Assuming the other problems are resolved, I'll close this issue, feel free to reopen!