Missing contig after merge #334
Labels
troubleshooting
workflow and data preparation questions
Comments
|
Re-running this just to double check. |
|
Hello @Ge0rges, If you can get a small exampe that will reproduce the bug, send me an email? |
|
Seems like I'm getting a lot of batch errors. Any tips for debugging? |
|
Send me the logs? |
|
On it. |
|
Log: Is this occurring perhaps because not all contigs occur in all bedmethyls? |
|
Tabix are generated with |
|
Maybe, let me look into it - sorry about the delay. |
|
Your standards for what constitutes a delay are too high 😂. Thanks for your help, hope I haven't missed something obvious. |
|
Just trying to keep up with you! I think I found the bug. Try this build. This fix and some others will get released very soon. |
|
Thanks! Testing it. |
|
Hi Art, That seems better (in the sense that the resulting bed is larger so presumably is skipping less data, and one type of error has disappeared) but similar errors remain in the log. I'm going to regenerate the tabix files and run again on my end. |
|
Hello @Ge0rges, Log lines such as this: Would be expected if there aren't bedMethyl records for that contig in any of the input bedMethyl files. Could you make sure that's the case? |
|
Seems like that is indeed the case, I ran a grep search on 4 different contigs. I guess they're present in the reference but no data. |
|
One last question @ArtRand , is it usual for |
|
Going back to the missing contig issues, wouldn't merge only look for contigs that are present in the tabix headers generated from the beds? So what's the case where it exists in the header but not the bed (which would result in these warnings?) Or is it the case that it looks in |
|
Hello @Ge0rges, That memory usage seems excessive. Let me do some stress tests on my side. In the mean time you could try reducing the
This is the case. The routine first looks grabs all the contigs from |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi @ArtRand,
I did
modkit bedmethyl merge --threads 40 --io-threads 10 -o all_samples.bed -g sizes.tsv *.bed.gz. I checked for example thatcontig_1is present insizes.tsvand inbarcode01.bed.gzbut not inall_samples.bed.No errors running
modkit bedmethyl merge.The text was updated successfully, but these errors were encountered: