bug: get_back_counts_for_line_plotf & get_back_res_meeting_min_occ

[dvc28]

Good day, I read your recent paper Zamkovaya et al 2021 in ISME and I am excited to implement a similar network analysis (mdmnets package) with my data, but I am running into issues at the start of the code you provide. I have 16S amplicon data analysed with dada2 pipeline (ASV table) for 23 samples and 70689 taxa in total.

I can create my phyloseq object without problems

seqtab.nochim<-readRDS("~/seqtab.nochim_ice8.rds")
rep.seqs <- colnames(seqtab.nochim)
rep.seqs <- Biostrings::DNAStringSet(rep.seqs)
otu.names <- RDPutils::make_otu_names(1:length(rep.seqs))
colnames(seqtab.nochim) <- otu.names
names(rep.seqs) <- otu.names
asv.table <- otu_table(seqtab.nochim, taxa_are_rows=FALSE)
phylo <- phyloseq(asv.table, rep.seqs) 

then I can estimate the prevalence, no problem

prevdf = apply(X = otu_table(phylo),MARGIN = ifelse(taxa_are_rows(phylo), yes=1, no=2), FUN=function(x){sum(x>0)}) 
  prevdf = data.frame(Prevalence = prevdf, TotalAbundance = taxa_sums(phylo))
  prevdf <- prevdf[prevdf$Prevalence > 0,]

but when I run this line prev_lineplot <- get_back_counts_for_line_plotf (prevdf, "Environment") Error in $<-.data.frame(tmp, "Rank6", value = character(0)) : replacement has 0 rows, data has 70689 I cannot make sense of this error. I also added metadata (sample_table) and taxonomy (taxa_table) to the phyloseq object but I still get this error.

and when I try

phylo_for_net <- get_back_res_meeting_min_occ(phylo, filter_val_percent=0.4)

Error in validObject(.Object) : invalid class “otu_table” object: OTU abundance data must have non-zero dimensions. I have been checking my otu_table and I cannot find any obvious errors in my phyloseq object, all taxa and samples have values>1. I have tried pre-filtering my otu_table to eliminate low abundance and singletons with

filterlist  <- filterfun(kOverA(k=1, A=1, na.rm=FALSE))   
phylo  <- filter_taxa(phylo, filterlist, TRUE)

but I still get the same error with "get_back_res_meeting_min_occ". I have used this same data set to run Deseq2 and that worked smoothly so I am not sure what is the problem here.

I am very grateful for any feedback you may have regarding these issues. Thank you for your time.

[nicolereynolds1]

I am having a similar issue as well. I would really like to use these analyses on my data, but I haven't been able to get past these errors.

I can make my phyloseq object and do the prevalence calculations, but at the next step, I get this error, same as dvc28: prev_lineplot <- get_back_counts_for_line_plotf(prevdf, "bac+fungi") Error in$<-.data.frame(tmp, "Rank6", value = "MDM") : replacement has 1 row, data has 0

Also in the next step I get another error: phylo_for_net <- get_back_res_meeting_min_occ(phylo, filter_val_percent=0.4) Error in access(object, "sam_data", errorIfNULL) : sam_data slot is empty

Any help would be greatly appreciated as I am excited to get results with this method for my data.

Thank you!