Suggestion count data file filtering

[desmodus1984]

Hi,
I was finally able to create proper and accepted input files to run CAFE5.

cafe5 -i Test.CAFE5.tsv -t tree13_ultra.nwk

I was reading through the tutorial and I found a recommendation to filter the gene count input file,
I downloaded the python script, I run Linux Ubuntu 24.04, and I got an error:

python clade_and_size_filter.py -i Test.CAFE5.tsv -o filtered_Test.CAFE5.tsv -s
Traceback (most recent call last):
  File "/home/jpac1984/Fam-Evol/clade_and_size_filter.py", line 129, in <module>
    lines_to_keep_set = clade_filter(args.input_file, args.clade_str)
                        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/jpac1984/Fam-Evol/clade_and_size_filter.py", line 49, in clade_filter
    clade_count = sum(1 for sp_count in spp_info if int(sp_count) >= 1)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/jpac1984/Fam-Evol/clade_and_size_filter.py", line 49, in <genexpr>
    clade_count = sum(1 for sp_count in spp_info if int(sp_count) >= 1)
                                                    ^^^^^^^^^^^^^
ValueError: invalid literal for int() with base 10: 'Arcgaz'

I was suprised that the filtering failed but the analysis ran fine:

cat Base_results.txt
Model Base Final Likelihood (-lnL): 205518
Lambda: 0.0021671572091251
Maximum possible lambda for this topology: 0.0233981
56 values were attempted (0% rejected)

I used Orthofinder to find the Orthogroups and then modified the *.GeneCount.tsv in Rstudio to create input file for CAFE5. Perhaps is there any criterion that I can/should use to remove "outliers" or "weird" Orthogroups in R?

Thanks

[hahnlab-user]

I believe that python script was written using python2, which is likely why there was an error. Otherwise, looking for families with large variances (if this is an error you're getting from CAFE) usually solves the problem.