checkm tree_qa throws AttributeError

[aistBMRG]

Hi,

I am using checkM version 1.0.3. I installed all yesterday, ran the "checkm test" and no issues came up.

Now, I got an error when analyzing my own genome bins, as explained below:

command 1: checkm tree -t 20 ./ ./checkm_tree -x .fa

completed with no issues

command 2: checkm tree_qa -f ./checkm_tree.txt ./checkm_tree

error

---

[CheckM - tree_qa] Assessing phylogenetic markers found in each bin.

---

Reading HMM info from file. Parsing HMM hits to marker genes: Finished parsing hits for 3600 of 3600 (100.00%) bins.

Unexpected error: <type 'exceptions.AttributeError'> Traceback (most recent call last): File "/srv/sw/checkm/1.0.3/bin/checkm", line 709, in checkmParser.parseOptions(args) File "/srv/sw/checkm/1.0.3/lib/python2.7/site-packages/checkm/main.py", line 1197, in parseOptions self.treeQA(options) File "/srv/sw/checkm/1.0.3/lib/python2.7/site-packages/checkm/main.py", line 177, in treeQA treeParser.printSummary(options.out_format, options.tree_folder, RP, options.bTabTable, options.file, binStats) File "/srv/sw/checkm/1.0.3/lib/python2.7/site-packages/checkm/treeParser.py", line 45, in printSummary self.reportBinTaxonomy(outDir, resultsParser, bTabTable, outFile, binStats, bLineageStatistics=False) File "/srv/sw/checkm/1.0.3/lib/python2.7/site-packages/checkm/treeParser.py", line 633, in reportBinTaxonomy binIdToTaxonomy = self.getBinTaxonomy(outDir, binIds) File "/srv/sw/checkm/1.0.3/lib/python2.7/site-packages/checkm/treeParser.py", line 216, in getBinTaxonomy domainNode = self.findDomainNode(node) File "/srv/sw/checkm/1.0.3/lib/python2.7/site-packages/checkm/treeParser.py", line 237, in findDomainNode curNode = curNode.parent_node

AttributeError: 'NoneType' object has no attribute 'parent_node'

Would it be possible to clarify? Maybe I am running the commands incorrectly or my installation is faulty ...

Thank you very much in advance for your help.

Dieter.

[donovan-h-parks]

Hello Dieter,

It looks like you are running CheckM correctly. Perhaps this is a bug that we have not encountered. I noticed that you are running a lot of genomes through CheckM (3600!). Can you try running just a few genomes (say, 2 or 3) to see if the problem persists? If so, it should be easier to determine what is happening.

Cheers, Donovan

[aistBMRG]

Hi Donovan,

Thank you for the quick response. Yes, I used many bins ... I guess I was not expecting any trouble due to the number of bins.

I tried your suggestion of running with less bins/genomes (I used 2) and now it completed without error. Not sure what is happening or at what point (how many genomes) the process gives the error ...

At this point I will run in smaller batches and remove some bins.

Thanks again !!!

Dieter.

[donovan-h-parks]

The number of bins shouldn't matter. My guess is one of the bins is problematic. If you can identify which bin it is, I can determine why it is causing a problem.

[aistBMRG]

Ok, I will run a loop and then see which one fails. Btw, I just tested a subset of 504 and no issues ...

[donovan-h-parks]

I am assuming this can be closed as there has been no follow-up in over a month.