Make Heatmap/dendogram figure just from ANI matrix table?

[TonyMannion]

I have an ANI matrix table that I calculated using a different program, but I like the heatmap/dendogram output generated by pyani. Is it possible to use my ANI results to create a heatmap/dendogram figure using pyani code?

Thank you!

Tony

[widdowquinn]

Hi Tony,

If it's named appropriately (with corresponding class and label files as appropriate), you could drop it into a results directory and use the --rerender options.

If you're comfortable in Python you could import pyani as a package in Python, load your matrix as a pandas dataframe, and use the pyani_graphics module functions, e.g. call heatmap_seaborn().

Cheers,

L.

[TonyMannion]

I ran the below command with known data and got expected results. average_nucleotide_identity.py -v -i Contigs -o Genomes_ANIb_output/ -g --gmethod seaborn --gformat png,pdf,eps,svg --write_excel -m ANIb However, I got the error below when I then changed one value in the ANIb_percentage_identity.tab output and ran the command with --rerender average_nucleotide_identity.py -v -i Contigs -o Genomes_ANIb_output/ -g --gmethod seaborn --gformat png,pdf,eps,svg --write_excel -m ANIb --rerender INFO: pyani version: 0.2.3 INFO: Namespace(blastall_exe='blastall', blastn_exe='blastn', classes=None, force=False, formatdb_exe='formatdb', fragsize=1020, gformat='png,pdf,eps,svg', gmethod='seaborn', graphics=True, indirname='Contigs', jobprefix='ANI', labels=None, logfile=None, makeblastdb_exe='makeblastdb', maxmatch=False, method='ANIb', noclobber=False, nocompress=False, nucmer_exe='nucmer', outdirname='Genomes_ANIb_output/', rerender=True, scheduler='multiprocessing', seed=None, sgegroupsize=10000, skip_blastn=False, skip_nucmer=False, subsample=None, verbose=True, workers=None, write_excel=True) INFO: command-line: /usr/local/bin/average_nucleotide_identity.py -v -i Contigs -o Genomes_ANIb_output/ -g --gmethod seaborn --gformat png,pdf,eps,svg --write_excel -m ANIb --rerender INFO: Input directory: Contigs INFO: Removing directory Genomes_ANIb_output/ and everything below it WARNING: NOCLOBBER: not actually deleting directory INFO: Creating directory Genomes_ANIb_output/ INFO: NOCLOBBER+FORCE: not creating directory INFO: Output directory: Genomes_ANIb_output/ INFO: Using ANI method: ANIb WARNING: --rerender option used WARNING: Producing graphics with no new recalculations INFO: Rendering output graphics INFO: Formats requested: png,pdf,eps,svg INFO: Graphics format: png INFO: Graphics method: seaborn Traceback (most recent call last): File "/usr/local/bin/average_nucleotide_identity.py", line 791, in <module> draw(methods[args.method][1], gfmt, args.rerender) File "/usr/local/bin/average_nucleotide_identity.py", line 620, in draw logger.info("Writing heatmap to %s", outfilename) AttributeError: 'bool' object has no attribute 'info' Any ideas how to fix this? Thank you! Tony

…

On Mon, Jun 5, 2017 at 11:29 AM, Leighton Pritchard < ***@***.***> wrote: Hi Tony, If it's named appropriately (with corresponding class and label files as appropriate), you could drop it into a results directory and use the --rerender options. If you're comfortable in Python you could import pyani as a package in Python, load your matrix as a pandas dataframe, and use the pyani_graphics module functions, e.g. call heatmap_seaborn(). Cheers, L. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#85 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/Ab2jOWXQmorC-kZMQIxQn51FwNtlPoHmks5sBB7OgaJpZM4NwLJK> .

[widdowquinn]

Would you be able to provide the .tab file before and after the change you made?

[TonyMannion]

Before and after files are attached. Note I only changed the ANIb_percentage_identity.tab file and not the other .tab outputs such for alignment coverage. The before and after files had the same name: ANIb_percentage_identity.tab

…

On Mon, Jun 5, 2017 at 2:30 PM, Leighton Pritchard ***@***.*** > wrote: Would you be able to provide the .tab file before and after the change you made? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#85 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/Ab2jObnCUHhJkJkUkFGSMSvlOpm_s7pFks5sBElHgaJpZM4NwLJK> .

[widdowquinn]

I can't see any files on the thread. Do you see them?

[widdowquinn]

I think you sent your last comment by email. This strips out attachments. Would you be able to drag-and-drop the files into this thread on GitHub, please?

[TonyMannion]

I did attach them in my email response, sorry about that! Now attached in the zip folder.

Mannion_GitHub.zip

[widdowquinn]

Hi Tony - pyani requires all the .tab files for a rerender - could you please attach a .zip with all the relevant output?

[widdowquinn]

For info, I've attempted to reproduce this with the test data provided in the repository.

First, I generated specific output data:

average_nucleotide_identity.py -i tests/test_ani_data/ -o tests/issue_85 -m ANIm -g --gmethod seaborn -v -l issue_85.log

Then copied this to a new location:

cp -R tests/issue_85 tests/issue_85_after

I modified the ANIm_percentage_identity.tab data in a similar way to your modification, replacing 1.0 values on the diagonal with 0.5 (this makes no biological sense, BTW - it implies a sequence has only 50% identity with itself), and then ran the rerender:

average_nucleotide_identity.py -i tests/test_ani_data/ -o tests/issue_85_after -m ANIm -g --gmethod seaborn -v -l issue_85_rerender.log --rerender

This gave the same error you see:

INFO: Graphics method: seaborn
Traceback (most recent call last):
  File "/Users/lpritc/Documents/Development/pyani/venv-pyani-devel/bin/average_nucleotide_identity.py", line 4, in <module>
    __import__('pkg_resources').run_script('pyani==0.2.3', 'average_nucleotide_identity.py')
  File "/Users/lpritc/Documents/Development/pyani/venv-pyani-devel/lib/python3.6/site-packages/pkg_resources/__init__.py", line 738, in run_script
    self.require(requires)[0].run_script(script_name, ns)
  File "/Users/lpritc/Documents/Development/pyani/venv-pyani-devel/lib/python3.6/site-packages/pkg_resources/__init__.py", line 1506, in run_script
    exec(script_code, namespace, namespace)
  File "/Users/lpritc/Documents/Development/pyani/venv-pyani-devel/lib/python3.6/site-packages/pyani-0.2.3-py3.6.egg/EGG-INFO/scripts/average_nucleotide_identity.py", line 791, in <module>
  File "/Users/lpritc/Documents/Development/pyani/venv-pyani-devel/lib/python3.6/site-packages/pyani-0.2.3-py3.6.egg/EGG-INFO/scripts/average_nucleotide_identity.py", line 620, in draw
AttributeError: 'bool' object has no attribute 'info'

But so does the unmodified output:

average_nucleotide_identity.py -i tests/test_ani_data/ -o tests/issue_85 -m ANIm -g --gmethod seaborn -v -l issue_85_rerender.log --rerender

So this is an issue with the rerender code in general. I'll get onto it.

[widdowquinn]

OK - I've found the issue, and I'll push the change in a few minutes.

Before:

After:

[widdowquinn]

That's the fix implemented with commit 99bb596 - please could you clone the latest version, install, and check it works for you?

[TonyMannion]

Works perfectly now! Thank you so much!

Best,
Tony

[widdowquinn]

No worries - please raise another issue if you get any problems.

L.