ValueError in PF (single block)

[Tagliacollo]

(config file in Crawford_2012/PF-mp-one_subset_aln)

I received the following message when running PF with single block:

Victor-A-Tagliacollo-MacBook-Pro:~ Tagliacollo$ python2 /Applications/partitionfinder-2.1.1/PartitionFinder.py /Users/Tagliacollo/Desktop/Test --raxml --no-ml-tree --rcluster-percent 10 --weights '1, 0, 0, 0'
INFO | 2016-12-19 15:53:36,381 | ------------- PartitionFinder 2.1.1 -----------------
INFO | 2016-12-19 15:53:36,381 | You have Python version 2.7
INFO | 2016-12-19 15:53:36,381 | Command-line arguments used: /Applications/partitionfinder-2.1.1/PartitionFinder.py /Users/Tagliacollo/Desktop/Test --raxml --no-ml-tree --rcluster-percent 10 --weights 1, 0, 0, 0
INFO | 2016-12-19 15:53:36,382 | ------------- Configuring Parameters -------------
INFO | 2016-12-19 15:53:36,382 | Setting datatype to 'DNA'
INFO | 2016-12-19 15:53:36,382 | Setting phylogeny program to 'raxml'
INFO | 2016-12-19 15:53:36,382 | Program path is here /Applications/partitionfinder-2.1.1/programs
INFO | 2016-12-19 15:53:36,383 | Setting cluster_weights to: subset_rate = 1, freqs = 0, model = 0, alpha 0
INFO | 2016-12-19 15:53:36,383 | Setting working folder to: '/Users/Tagliacollo/Desktop/Test'
INFO | 2016-12-19 15:53:36,384 | Loading configuration at './partition_finder.cfg'
INFO | 2016-12-19 15:53:36,392 | Setting 'alignment' to 'Crawford_2012.phy'
INFO | 2016-12-19 15:53:36,393 | Setting 'branchlengths' to 'linked'
INFO | 2016-12-19 15:53:36,393 | You set 'models' to: GTR+G
INFO | 2016-12-19 15:53:36,404 | This analysis will use the following 1 models of molecular evolution
INFO | 2016-12-19 15:53:36,404 | GTR+G
INFO | 2016-12-19 15:53:36,405 | Setting 'model_selection' to 'aicc'
INFO | 2016-12-19 15:53:36,515 | Setting 'search' to 'rclusterf'
INFO | 2016-12-19 15:53:36,516 | ------------------------ BEGINNING NEW RUN -------------------------------
INFO | 2016-12-19 15:53:36,516 | Looking for alignment file './Crawford_2012.phy'...
INFO | 2016-12-19 15:53:36,522 | Using 4 cpus
INFO | 2016-12-19 15:53:36,522 | Beginning Analysis
INFO | 2016-12-19 15:53:36,674 | Checking previously run configuration data...
INFO | 2016-12-19 15:53:36,857 | Reading alignment file './Crawford_2012.phy'
INFO | 2016-12-19 15:53:36,939 | Reading alignment file './analysis/start_tree/source.phy'
INFO | 2016-12-19 15:53:37,265 | Starting tree file found.
INFO | 2016-12-19 15:53:37,265 | Performing relaxed clustering analysis
INFO | 2016-12-19 15:53:37,265 | Set rcluster-max to 1000
INFO | 2016-12-19 15:53:37,265 | PartitionFinder will have to analyse 0 subsets to complete this analyses
INFO | 2016-12-19 15:53:37,266 | *** Analysing starting scheme ***
INFO | 2016-12-19 15:53:37,382 | Finished subset 1/1, 100.00 percent done
INFO | 2016-12-19 15:53:37,400 | *** Relaxed clustering algorithm step 1 of up to 0 ***
INFO | 2016-12-19 15:53:37,400 | Measuring the similarity of 0 subset pairs
Traceback (most recent call last):
File "/Applications/partitionfinder-2.1.1/PartitionFinder.py", line 25, in
sys.exit(main.main("PartitionFinder", "DNA"))
File "/Applications/partitionfinder-2.1.1/partfinder/main.py", line 401, in main
run_analysis(cfg, options)
File "/Applications/partitionfinder-2.1.1/partfinder/main.py", line 335, in run_analysis
results = anal.analyse()
File "/Applications/partitionfinder-2.1.1/partfinder/analysis.py", line 93, in analyse
self.do_analysis()
File "/Applications/partitionfinder-2.1.1/partfinder/logtools.py", line 159, in indented_fn
fn(*args, **kwargs)
File "/Applications/partitionfinder-2.1.1/partfinder/analysis_method.py", line 387, in do_analysis
the_config.cluster_weights)
File "/Applications/partitionfinder-2.1.1/partfinder/neighbour.py", line 118, in get_distance_matrix
final_dists = get_manhattan_matrix(rates, freqs, model, alpha, weights)
File "/Applications/partitionfinder-2.1.1/partfinder/neighbour.py", line 62, in get_manhattan_matrix
if np.amax(r_dists)>0:
File "/Users/Tagliacollo/anaconda2/lib/python2.7/site-packages/numpy/core/fromnumeric.py", line 2268, in amax
out=out, keepdims=keepdims)
File "/Users/Tagliacollo/anaconda2/lib/python2.7/site-packages/numpy/core/_methods.py", line 26, in _amax
return umr_maximum(a, axis, None, out, keepdims)
ValueError: zero-size array to reduction operation maximum which has no identity

[natalie-vandepol]

Did you ever figure this out? I am working on a single block amino acid set and have the same error.

[Tagliacollo]

Hi @natalie-vandepol,

This error is probably related to empty arrays because you have 0 subset pairs to compare similarity (single block). You can find the likelihood values of a single block in the output of PartitionFinder2 Have you posted this error in the PartitionFinder group? (https://groups.google.com/forum/#!forum/partitionfinder)

[roblanf]

Hi Natalie, Please post this on the partitionfinder google group. But a simple fix is not to use the rcluster algorithm - if you are looking at a single block then there is no clustering to be done. Cheers, Rob

…

On 23 September 2017 at 21:36, Victor A Tagliacollo < ***@***.***> wrote: Hi @natalie-vandepol <https://github.com/natalie-vandepol>, This error is probably related to empty arrays because you have 0 subset pairs to compare similarity (single block). You can find the likelihood values of a single block in the output of PartitionFinder2 Have you posted this error in the PartitionFinder group? (https://groups.google.com/ forum/#!forum/partitionfinder) — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#30 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA2pE1DJ1yb9F9FgGxaFekRRvTQnDrYmks5slO0vgaJpZM4LQX5z> .

-- Rob Lanfear Division of Ecology and Evolution, Research School of Biology, The Australian National University, Canberra www.robertlanfear.com