Closed LaserKate closed 7 years ago
joey711
commented
7 years ago This looks like an issue with the latest update in ggplot2. Can you share a reproducible example that generates this error? Perhaps with the included example dataset GlobalPatterns?
thanks
LaserKate
commented
7 years ago Hi Paul, I finally found a fix to this problem. I was trying to run the older version of Phyloseq with the newest version of R. I can get everything to work nicely if I use an older version of R and Phyloseq or alternatively, the newest version of Phyloseq with the newest version of R. sorry for the hassle, cheers, Kate
On Tue, Nov 8, 2016 at 4:19 AM, Paul J. McMurdie notifications@github.com wrote:
This looks like an issue with the latest update in ggplot2. Can you share a reproducible example that generates this error? Perhaps with the included example dataset GlobalPatterns?
thanks
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/joey711/phyloseq/issues/673#issuecomment-258917070, or mute the thread https://github.com/notifications/unsubscribe-auth/AKqtEg5YZgRvUruZqKqpz2Mcpi9hkcb5ks5q72umgaJpZM4KQhcF .
Kate Quigley Cuerpo de Paz-Peru Voluntaria de Medio Ambiente 995974412 RPM:*377992
elsherbini
commented
7 years ago I am having the same issue for plot_bar and for plot_heatmap:
R version 3.3.2 (2016-10-31) -- "Sincere Pumpkin Patch"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)
>library(phyloseq)
> packageVersion('phyloseq')
[1] ‘1.19.1’
> packageVersion('ggplot2')
[1] ‘2.2.0’
> #from http://joey711.github.io/phyloseq/plot_bar-examples.html
> data("GlobalPatterns")
> gp.ch = subset_taxa(GlobalPatterns, Phylum == "Chlamydiae")
> plot_bar(gp.ch)
Error: evaluation nested too deeply: infinite recursion / options(expressions=)?
> # from http://joey711.github.io/phyloseq/plot_heatmap-examples.html
> gpt <- subset_taxa(GlobalPatterns, Kingdom=="Bacteria")
> gpt <- prune_taxa(names(sort(taxa_sums(gpt),TRUE)[1:300]), gpt)
> plot_heatmap(gpt, sample.label="SampleType")
Error: evaluation nested too deeply: infinite recursion / options(expressions=)?If I increase the options(expressions), i get a protect() error for stack overflow:
> options("expressions"=20000)
> plot_heatmap(gpt, sample.label="SampleType")
Error: protect(): protection stack overflowI've set my stack size to be unlimited with ulimit -s unlimited
Downgrading ggplot2 to 2.1.0 did not fix the issue.
> packageVersion('ggplot2')
[1] ‘2.1.0’
> plot_bar(gpt)
Error: evaluation nested too deeply: infinite recursion / options(expressions=)?I'm afraid that I cannot reproduce your error. Can you give a different example that does reproduce the error, or more system information if you believe it is a system-dependent error?
Also, note that you are on the development branch of bioconductor, or at least are using the development release of phyloseq according to your version number. The version of phyloseq indicated in my log of your example, below, shows the current release version of phyloseq.
I will close for now, as OP said this was resolved by updating R and packages, and your example of an error appears to run fine. Please feel free to post if this is actually a persisting issue, and I will re-open. The current recommendation is to use the release version of phyloseq.
R.version## _
## platform x86_64-apple-darwin13.4.0
## arch x86_64
## os darwin13.4.0
## system x86_64, darwin13.4.0
## status
## major 3
## minor 3.2
## year 2016
## month 10
## day 31
## svn rev 71607
## language R
## version.string R version 3.3.2 (2016-10-31)
## nickname Sincere Pumpkin Patchlibrary("phyloseq"); packageVersion("phyloseq")##
## This data.table install has not detected OpenMP support. It will work but slower in single threaded mode.## [1] '1.18.1'packageVersion('ggplot2')## [1] '2.2.0'data("GlobalPatterns")
gp.ch = subset_taxa(GlobalPatterns, Phylum == "Chlamydiae")
plot_bar(gp.ch)
gpt <- subset_taxa(GlobalPatterns, Kingdom=="Bacteria")
gpt <- prune_taxa(names(sort(taxa_sums(gpt),TRUE)[1:300]), gpt)
plot_heatmap(gpt, sample.label="SampleType")## Warning: Transformation introduced infinite values in discrete y-axis
elsherbini
commented
7 years ago Hmm, I am running this stuff on my personal computer. I just restarted and installed the current release version of phyloseq. I have the same problem with that release (though if I don't set the stack size to unlimited I get a stack size error rather than infinite recursion error.
I'll let you know if I find a solution.
> R.version
_
platform x86_64-pc-linux-gnu
arch x86_64
os linux-gnu
system x86_64, linux-gnu
status
major 3
minor 3.2
year 2016
month 10
day 31
svn rev 71607
language R
version.string R version 3.3.2 (2016-10-31)
nickname Sincere Pumpkin Patch
> library("phyloseq"); packageVersion("phyloseq")
[1] ‘1.18.1’
> packageVersion('ggplot2')
[1] ‘2.2.0’
> data("GlobalPatterns")
> gp.ch = subset_taxa(GlobalPatterns, Phylum == "Chlamydiae")
> plot_bar(gp.ch)
Error: C stack usage 7969476 is too close to the limit
> gpt <- subset_taxa(GlobalPatterns, Kingdom=="Bacteria")
> gpt <- prune_taxa(names(sort(taxa_sums(gpt),TRUE)[1:300]), gpt)
> plot_heatmap(gpt, sample.label="SampleType")
Error: C stack usage 7969476 is too close to the limitis this a resource-constrained machine? That limit, if memory, looks to be very small...
elsherbini
commented
7 years ago That is the stack size, and I think that size is the default for most linux systems. I have 16GB RAM, does that seem like enough for the example?
On Thu, Dec 1, 2016 at 8:47 PM, Paul J. McMurdie notifications@github.com wrote:
is this a resource-constrained machine? That limit, if memory, looks to be very small...
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/joey711/phyloseq/issues/673#issuecomment-264351114, or mute the thread https://github.com/notifications/unsubscribe-auth/ACkgvRQfvyvWJTyyRC8hSGvuQIK4SEsDks5rD3i3gaJpZM4KQhcF .
elsherbini
commented
7 years ago As an update, this does appear to be a system issue of some sort. Still haven't tracked it down.
On the cluster I have access to I can produce the bar plot. On my personal computer I cannot, though I can produce ordination plots and diversity estimate plots.
joey711
commented
7 years ago Thanks for the update. I think this is out of the scope of the phyloseq issue tracker unless it appears to be more pervasive. If I'm remembering correctly, these plots are even checked in phyloseq's unit tests, which are run on all three of Bioconductor's testing servers before release (Ubuntu, MacOS, Win)... 16GB RAM should be plenty of total RAM, so my guess is it is a setting for your OS that has limited the stack to something that honestly seems a bit small, and this error is probably not the only reason you'd want to increase it.
Hope you solve the problem! For what it is worth, I do A LOT of my work on an Amazon AWS instance running Ubuntu and RStudio Server, even though I'm happy with the performance of my Macbook Pro. Pretty much the same user interface, so not a bad option at all.
ZellaD
commented
7 years ago elsherbini did you ever figure out how to change the C stack limit? I am running into the same problem and not sure how to change it. Everything online says to change it with ulimit -s but this has not worked for me. Any suggestions?
elsherbini
commented
7 years ago I never did solve it on that machine. It works fine on the cluster I have access to. I'm convinced that there is another silent error that leads to the infinite recursion problem, and so it's difficult to track down whats going wrong. Even with an unlimited stack size the problem isn't fixed, so I don't think that's the real issue.
apascualgarcia
commented
6 years ago Hi there, I have the same issue running in a 8GB laptop with Ubuntu 16, and it happens to me with plot_bar() but also using as.character() with tax_table() (see below). I thought that the problem was rstudio, because I'm running an old version (1.0.43), but then I updated the packages in my machine (excluding r studio and R, the latter is in the latest version) and the problem dissapears. Then I'm a little bit confused because with the problem present both R and rstudio were in the same version that the one I'm running now without problems, so it may be any library causing the problem that I now updated. I think this issue
may also be related so merging them would help (I found the other issue first because the error is explicit in the title).
> library(phyloseq)
> data("GlobalPatterns")
> gp.ch = subset_taxa(GlobalPatterns, Phylum == "Chlamydiae")
> plot_bar(gp.ch)
Error: C stack usage 7970212 is too close to the limit
> topESVs = names(sort(taxa_sums(gp.ch), TRUE)[1:19])
> taxTop = cbind(tax_table(gp.ch), family = NA, genus=NA)
> taxTop[topESVs, "family"] <- as(tax_table(taxTop)[topESVs, "Family"],
... "character") #
Error: C stack usage 7970372 is too close to the limit
> R.version
_
platform x86_64-pc-linux-gnu
arch x86_64
os linux-gnu
system x86_64, linux-gnu
status
major 3
minor 4.4
year 2018
month 03
day 15
svn rev 74408
language R
version.string R version 3.4.4 (2018-03-15)
nickname Someone to Lean On
> packageVersion('phyloseq')
[1] ‘1.22.3’
> packageVersion('ggplot2')
[1] ‘3.0.0’
Hello, I recently updated R and am now running R version 3.3.1 (2016-06-21).
However, I am still working with
phyloseq_1.10.0andggplot2_2.1.0Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows 7 x64 (build 7601) Service Pack 1I am trying to reproduce an old figure that used to work, but now get the following error when using the function plot_bar or plot_tree with a phyloseq object:
plot_bar(phylo, fill="Species")+theme_classic()Other functions work with the phylo object, namely, plot_ordination(). However,
Plot_tree also ends in the same error if I just write
plot_tree(phylo)orplot_tree(phylo, label.tips="taxa_names")But works correctly withplot_tree(phylo, "treeonly")Any idea how to fix this issue? I think it has something to do with updating R but some old packages dependencies are not properly communicating with phyloseq anymore...
Thanks in advance! Kate