search

HCBravoLab / metagenomeSeq

Statistical analysis for sparse high-throughput sequencing
66 stars 20 forks source link

#Error in (function (classes, fdef, mtable) : unable to find an inherited method for function ‘assayData’ for signature ‘"fitZigResults"’ #79

Open DitteOL opened 4 years ago

DitteOL commented 4 years ago

Hi,

I have a problem when comparing two levels in a given Rank using metagenomeseq. This have worked for me not long ago and now I only receive errors:

I use the function (model being the variable in which I want to compare the two levels):

metagenomeSeq_treatment_on_any_level_eff_Model = function(physeq,level) {
  require("metagenomeSeq")
  require("phyloseq")
  # Enforce orientation
  if (!taxa_are_rows(physeq)) {
    physeq <- t(physeq)
  }

o = as(otu_table(physeq), "matrix")
s<-as(data.frame(sample_data(physeq)), "AnnotatedDataFrame")# Convert sample_data to AnnotatedDataFrame
t<-as(data.frame(tax_table(physeq)), "AnnotatedDataFrame")# Convert tax_data to AnnotatedDataFrame
mr = newMRexperiment(o,phenoData=s,featureData=t)  # Create the metagenomeSeq object

mr_agg<-aggregateByTaxonomy(mr,lvl=level,norm=T,aggfun=colSums, log=F, sl=1000, out ="MRexperiment")
mr_agg<-cumNorm(mr_agg)  # Trigger metagenomeSeq to calculate its Cumulative Sum scaling factor.
treatm<-pData(mr_agg)$Model
mod = model.matrix(~treatm)
colnames(mod) = levels(treatm)
settings = zigControl(maxit=100,verbose=FALSE)
fit = fitZig(obj = mr_agg, mod=mod, control=settings)

a = MRfulltable(fit, number = nrow(assayData(fit)$counts), eff = .5)
b <- subset(a, adjPvalues<=0.05)

return(b)
}

I use a phyloseq object

phyloseq-class experiment-level object otu_table() OTU Table: [ 1254 taxa and 12 samples ] sample_data() Sample Data: [ 12 samples by 5 sample variables ] tax_table() Taxonomy Table: [ 1254 taxa by 7 taxonomic ranks ] phy_tree() Phylogenetic Tree: [ 1254 tips and 1253 internal nodes ]

in the function:

sig_lvl2_eff_NASH = metagenomeSeq_treatment_on_any_level_eff_Model(Ellegaard,"Rank2")

And gets this error code with debug:

Error in (function (classes, fdef, mtable) : unable to find an inherited method for function ‘assayData’ for signature ‘"fitZigResults"’

  1. stop(gettextf("unable to find an inherited method for function %s for signature %s", sQuote(fdef@generic), sQuote(cnames)), domain = NA)
  2. (function (classes, fdef, mtable) { methods <- .findInheritedMethods(classes, fdef, mtable) if (length(methods) == 1L) ...
  3. assayData(fit)
  4. nrow(assayData(fit)$counts)
  5. MRfulltable(fit, number = nrow(assayData(fit)$counts), eff = 0.5)
  6. metagenomeSeq_treatment_on_any_level_eff_Model(Ellegaard, "Rank2") 
sessionInfo()

R version 4.0.0 (2020-04-24) Platform: x86_64-apple-darwin17.0 (64-bit) Running under: macOS Catalina 10.15.4

Matrix products: default BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib LAPACK: /Library/Frameworks/R.framework/Versions/4.0/Resources/lib/libRlapack.dylib

locale: [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages: [1] stats4 grid parallel stats graphics grDevices utils datasets [9] methods base

other attached packages: DESeq2_1.28.1
SummarizedExperiment_1.18.1 DelayedArray_0.14.0
matrixStats_0.56.0
GenomicRanges_1.40.0
GenomeInfoDb_1.24.0
IRanges_2.22.2
S4Vectors_0.26.1
pheatmap_1.0.12
gtable_0.3.0
extrafont_0.17
extrafontdb_1.0
MASS_7.3-51.6
reshape2_1.4.4
gmodels_2.18.1
gtools_3.8.2
gplots_3.0.3
metagenomeSeq_1.30.0
[19] RColorBrewer_1.1-2
glmnet_4.0
Matrix_1.2-18
limma_3.44.1
Biobase_2.48.0
BiocGenerics_0.34.0
doParallel_1.0.15
iterators_1.0.12
foreach_1.5.0
vegan_2.5-6
lattice_0.20-41
permute_0.9-5
knitr_1.28
plyr_1.8.6
ggplot2_3.3.1
phyloseq_1.32.0

loaded via a namespace (and not attached): [1] nlme_3.1-148 bitops_1.0-6 bit64_0.9-7
[4] tools_4.0.0 R6_2.4.1 KernSmooth_2.23-17
[7] DBI_1.1.0 mgcv_1.8-31 colorspace_1.4-1
[10] ade4_1.7-15 withr_2.2.0 tidyselect_1.1.0
[13] bit_1.1-15.2 compiler_4.0.0 labeling_0.3
[16] caTools_1.18.0 scales_1.1.1 genefilter_1.70.0
[19] digest_0.6.25 stringr_1.4.0 XVector_0.28.0
[22] pkgconfig_2.0.3 rlang_0.4.6 rstudioapi_0.11
[25] RSQLite_2.2.0 farver_2.0.3 shape_1.4.4
[28] generics_0.0.2 jsonlite_1.6.1 BiocParallel_1.22.0
[31] dplyr_1.0.0 RCurl_1.98-1.2 magrittr_1.5
[34] GenomeInfoDbData_1.2.3 biomformat_1.16.0 Rcpp_1.0.4.6
[37] munsell_0.5.0 Rhdf5lib_1.10.0 ape_5.4
[40] lifecycle_0.2.0 stringi_1.4.6 zlibbioc_1.34.0
[43] rhdf5_2.32.0 blob_1.2.1 gdata_2.18.0
[46] crayon_1.3.4 Biostrings_2.56.0 splines_4.0.0
[49] multtest_2.44.0 annotate_1.66.0 locfit_1.5-9.4
[52] pillar_1.4.4 igraph_1.2.5 Wrench_1.6.0
[55] geneplotter_1.66.0 codetools_0.2-16 XML_3.99-0.3
[58] glue_1.4.1 BiocManager_1.30.10 data.table_1.12.8
[61] vctrs_0.3.1 Rttf2pt1_1.3.8 purrr_0.3.4
[64] xfun_0.14 xtable_1.8-4 survival_3.1-12
[67] tibble_3.0.1 AnnotationDbi_1.50.0 memoise_1.1.0
[70] cluster_2.1.0 ellipsis_0.3.1

Thank you for the much needed help!

hcorrada commented 4 years ago

We made a change to the representation of results (that’s what fitZigResults is). I’ll take a look to check if we missed an update to the MRfulltable method.

DitteOL commented 4 years ago

Dear hcorrada I was wandering if there was any progress on this issue? Thank in advance!

dombraccia commented 4 years ago

Hi @DitteOL , I’m looking into the problem. Would you mind sending me a subset of the phyloseq dataset you are working with? It would help me create a reproducible example. You can send it to dbraccia@umd.edu

dombraccia commented 4 years ago

@DitteOL could you update your version of metagenomeSeq to the most current? It seems you are running on version 1.30 and the most current version is 1.31.

DitteOL commented 4 years ago

I have tried this several times but I'm not allowed to go higher than v 1.30.

Thank you for your prompt response!

dombraccia commented 4 years ago

@DitteOL If you haven't already, could you try installing metagenomeSeq via: BiocManager::install("hcbravolab/metagenomeSeq") and then check the version to see if it updates?

DitteOL commented 4 years ago

Now I'm allowed to update matagenomeSeq. But I still get the same error when trying to compare two levels:

Error in (function (classes, fdef, mtable) : unable to find an inherited method for function ‘assayData’ for signature ‘"fitZigResults"’

Any other ideas? :)

DitteOL commented 4 years ago

@dombraccia After updating several times and are now at RStudio version 1.3.1056 and R version 4.0.2 (macOS Catalina) plus supplementing my code with:

no_cores <- detectCores() - 1 library(doParallel)

create the cluster for caret to use

cl <- makePSOCKcluster(no_cores)

cl <- parallel::makeCluster(no_cores, setup_strategy = "sequential") registerDoParallel(cl)

I get this error message:

Error in h(simpleError(msg, call)) : error in evaluating the argument 'x' in selecting a method for function 'nrow': unable to find an inherited method for function ‘assayData’ for signature ‘"fitZigResults"’

I was wondering if I'm back to the start or the problem have changed?

Again your help is much appreciated!

dombraccia commented 4 years ago

Hey @DitteOL,

It appears that you are still running into the same problem, since there is not a specific assayData generic method built in for the fitZigResults class. What I need to do on my end is test out an assayData method on this results class and that will likely solve all your problems.

However, I am curious about the latest code you have sent. I assume the doParallel code does not interfere with the metagenomeSeq_treatment_on_any_level_eff_Model function you have specified above? If not, then I will continue using the data you have sent me over email and the code above to try and debug the problem.