Differential abundance: failure with corral data

[asizemore]

From Dan B on live site

So far i know

• HMP WGS on qa works (sex as group comparison variable), but only two organisms have results
• Bonus computation fails on qa

Logs from the bonus failure:

[1] "starting differential abundance computation"

2024-01-08 17:30:41 Received abundance table with 1279 samples and 21 taxa.

2024-01-08 17:30:41 Found 1279 samples with a value for EUPATH_0000096.EUPATH_0000639 in either groupA or groupB. The calculation will continue with only these samples.
[1] "Creating output folder"
2024-01-08 17:30:41 INFO::Writing function arguments to log file
2024-01-08 17:30:41 INFO::Verifying options selected are valid
2024-01-08 17:30:41 INFO::Determining format of input files
2024-01-08 17:30:41 INFO::Input format is data samples as rows and metadata samples as rows
2024-01-08 17:30:41 INFO::Formula for fixed effects: expr ~  EUPATH_0000096.EUPATH_0000639
2024-01-08 17:30:41 INFO::Filter data based on min abundance and min prevalence
2024-01-08 17:30:41 INFO::Total samples in data: 1279
2024-01-08 17:30:41 INFO::Min samples required with min abundance for a feature not to be filtered: 127.900000
2024-01-08 17:30:41 INFO::Total filtered features: 18
2024-01-08 17:30:41 INFO::Filtered feature names from abundance and prevalence filtering: OBI_0002623.EUPATH_0009269_Ascomycota_Saccharomycetales_Debaryomycetaceae_Candida_Candida_orthopsilosis, OBI_0002623.EUPATH_0009269_Ascomycota_Saccharomycetes_Saccharomycetales_Candida_Candida_sake, OBI_0002623.EUPATH_0009269_Ascomycota_Saccharomycetes_Saccharomycetales_Debaryomycetaceae_, OBI_0002623.EUPATH_0009269_Ascomycota_accharomycetales_Metschnikowiaceae_Clavispora_Candida_intermedia, OBI_0002623.EUPATH_0009269_Ascomycota_aromycetales_Saccharomycodaceae_Hanseniaspora_Hanseniaspora_uvarum, OBI_0002623.EUPATH_0009269_Ascomycota_aromycetes_Saccharomycetales_Pichiaceae_Pichia_Pichia_kudriavzevii, OBI_0002623.EUPATH_0009269_Ascomycota_ccharomycetales_Metschnikowiaceae_Clavispora_Clavispora_lusitaniae, OBI_0002623.EUPATH_0009269_Ascomycota_ccharomycetales_Saccharomycetaceae_Nakaseomyces_Candida_glabrata, OBI_0002623.EUPATH_0009269_Ascomycota_cetales_Trichomonascaceae_Wickerhamiella_Wickerhamiella_pararugosa, OBI_0002623.EUPATH_0009269_Ascomycota_cetes_Saccharomycetales_Debaryomycetaceae_Candida_Candida_albicans, OBI_0002623.EUPATH_0009269_Ascomycota_haromycetales_Debaryomycetaceae_Debaryomyces_Debaryomyces_hansenii, OBI_0002623.EUPATH_0009269_Ascomycota_omycetes_Eurotiales_Aspergillaceae_Aspergillus_Aspergillus_sydowii, OBI_0002623.EUPATH_0009269_Ascomycota_s_Saccharomycetales_Debaryomycetaceae_Candida_Candida_parapsilosis, OBI_0002623.EUPATH_0009269_Ascomycota_tes_Saccharomycetales_Debaryomycetaceae_Candida_Candida_tropicalis, OBI_0002623.EUPATH_0009269_Ascomycota_ycetales_Saccharomycetaceae_Saccharomyces_Saccharomyces_cerevisiae, OBI_0002623.EUPATH_0009269_Basidiomycota_Sporidiobolales_Sporidiobolaceae_Rhodotorula_Rhodotorula_sp_JG_1b, OBI_0002623.EUPATH_0009269_Basidiomycota_cetes_Malasseziales_Malasseziaceae_Malassezia_Malassezia_restricta, OBI_0002623.EUPATH_0009269_Basidiomycota_mycetes_Malasseziales_Malasseziaceae_Malassezia_Malassezia_globosa
2024-01-08 17:30:41 INFO::Total filtered features with variance filtering: 0
2024-01-08 17:30:41 INFO::Filtered feature names from variance filtering:
2024-01-08 17:30:41 INFO::Running selected normalization method: TSS
2024-01-08 17:30:41 INFO::Applying z-score to standardize continuous metadata
2024-01-08 17:30:41 INFO::Running selected transform method: LOG
2024-01-08 17:30:41 INFO::Running selected analysis method: LM
Error in seq_len(ncol(features)) :
  argument must be coercible to non-negative integer

So that's the issue. The variance filter chops off the features.

Probably a good reason to prioritize adding a way to see the logs of each computation

[asizemore]

I read a bit more into the error and it's actually not the variance filter. The variance filter output is saying that only 0 samples were removed, not that we kept 0 samples after the filter. A little confusing. Anyways another person had a similar issue: https://forum.biobakery.org/t/error-in-seq-length-ncol-features-argument-must-be-coercible-to-non-negative-integer/5757

[asizemore]

Probably related to https://github.com/VEuPathDB/EdaNewIssues/issues/600

[d-callan]

confirmed working w microbiomeComputations v3.3.5