Issues with hclust2 operation and compatibilites

[Megan Folkerts]

Hello,

This morning, while trying to run hclust2, I was getting this error.

/cm/local/apps/slurm/var/spool/job1756051/slurm_script:152: FutureWarning: read_table is deprecated, use read_csv instead.

  index_col = self.args.sname_row if self.args.sname_row > -1 else None

Traceback (most recent call last):

  File "/cm/local/apps/slurm/var/spool/job1756051/slurm_script", line 800, in <module>

    cl.shcluster()

  File "/cm/local/apps/slurm/var/spool/job1756051/slurm_script", line 376, in shcluster

    self.shclusters = sph.linkage(self.s_dm, method=args.slinkage)

  File "/home/mfolkerts/miniconda3/envs/metaphlan2/lib/python2.7/site-packages/scipy/cluster/hierarchy.py", line 1109, in linkage

    raise ValueError("The condensed distance matrix must contain only "

ValueError: The condensed distance matrix must contain only finite values.

I've seen other, older posts that say that this is a compatibility issue, possibly with scipy. I downgraded scipy to 0.18.1, however the error persisted. Eventually, I completely removed my metaphlan2 conda environment and repeated the install. Now, however, I can't install hclust2 at all, as I now get the following error:

UnsatisfiableError: The following specifications were found to be incompatible with each other:

  - pip -> python[version='>=3.6,<3.7.0a0']

When I try to change to upgrade or downgrade python to a different version, I get more compatibility issues. Any help would be greatly appreciated at this point.

[Valentina Galata]

I have the same issue when trying to install the latest MetaPhlAn2 version (2.9.15) togther with hclust2. There are many conflicts for various packages.

The only thing that worked for me was using no version numbers and let conda decide what to install:

channels:
    - conda-forge
    - bioconda
dependencies:
    - metaphlan2
    - hclust2

Then, conda installed MetaPhlAn2 v. 2.9.14, hclust2 0.99 and scipy 1.2.1. However, I have not tested the tools yet.

[Francesco Beghini]

Hi Megan,

could you attach here the profile you tried to plot with hclust2?

I'll have a look and update also the hclust2 package.

Best,

Francesco Beghini

PhD Student - Laboratory of Computational Metagenomics

Department of Cellular, Computational and Integrative Biology - CIBIO

University of Trento

This electronic message is intended to be for the use only of the named recipient, and may contain information that is confidential or privileged, including patient health information. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution or use of the contents of this message is strictly prohibited. If you have received this message in error or are not the named recipient, please notify us immediately by contacting the sender at the electronic mail address noted above, and delete and destroy all copies of this message. Thank you.

--
You received this message because you are subscribed to the Google Groups "MetaPhlAn-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to metaphlan-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/metaphlan-users/c5bf2dea-d7e7-4ce7-a415-d28eefbfedae%40googlegroups.com.

[Megan Folkerts]

This is the merged abundance table that was giving me the issue. 

To view this discussion on the web visit https://groups.google.com/d/msgid/metaphlan-users/CALkF7uY4WVNbFGPW03AhhmXVx36vkMrUaTgH5DHCDUKUjn9f3Q%40mail.gmail.com.

--

Megan Folkerts, MS
Research Associate II
Center for Emerging Pathogens and Technologies
TGen North
Flagstaff, Arizona
928-226-6375

[Megan Folkerts]

I've been tinkering with this more, and I was eventually able to get a heatmap made. I had a sample in which nothing aligned (all values were 0) which turned out to be the reason for the initial error. Once I removed that sample, I started getting a new error, but in looking closely at the table, I saw that there were species that were duplicated several times. When both of these issues were corrected, a heat map was generated using commands similar to those in the tutorial. Just posting this in case others are getting hung up with these same errors.

[Francesco Beghini]

Hi Megan,

I re-wrote the merge_metaphlan_tables script and it is now available for MetaPhlAn2 2.9.19 both on the repo and in the conda repository (in a couple of hours).

Could you check wheather this update solved the issue?

Thanks,

Francesco Beghini

PhD Student - Laboratory of Computational Metagenomics

Department of Cellular, Computational and Integrative Biology - CIBIO

University of Trento

[Megan Folkerts]

Hi Francesco,

Yes, the issue seems to be resolved. Sorry for the delayed response.

Thank you!

Megan