Question

sc3_prepare function error

1

Entering edit mode

Assa Yeroslaviz ★ 1.5k

@assa-yeroslaviz-1597

Last seen 3 months ago

Germany

Hi,

I am trying to follow the workflow from the hemberg lab about single-cell analysis. There is though a conflict in the versions.

I try to run the SC3 package. There is a function here called sc3_prepare, but under my R environment it doesn't seem to exist.

I keep getting the error message (running the script in the file 17-clustering.Rmd)

Quitting from lines 94-95 (17-clustering.Rmd) 
Error in eval(expr, envir, enclos) : 
  could not find function "sc3_prepare"
Calls: <Anonymous> ... handle -> withCallingHandlers -> withVisible -> eval -> eval
In addition: Warning messages:
1: package 'mclust' was built under R version 3.3.2 
2: package 'ggplot2' was built under R version 3.3.2 
Execution halted

looking at my sc3 package i can't find this function at all.

is there something i am doing wrong here?

thanks

Assa

> sessionInfo()
R version 3.3.0 (2016-05-03)
Platform: x86_64-apple-darwin13.4.0 (64-bit)
Running under: OS X 10.11.6 (El Capitan)

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] parallel  stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] SC3_1.1.4            scater_1.0.4         ggplot2_2.2.1        Biobase_2.32.0       BiocGenerics_0.18.0 
[6] BiocInstaller_1.22.3

loaded via a namespace (and not attached):
 [1] bitops_1.0-6         matrixStats_0.51.0   doParallel_1.0.10    RColorBrewer_1.1-2   rprojroot_1.2       
 [6] tools_3.3.0          backports_1.0.5      doRNG_1.6            R6_2.2.0             KernSmooth_2.23-15  
[11] DBI_0.5-1            lazyeval_0.2.0       colorspace_1.3-2     gridExtra_2.2.1      pkgmaker_0.22       
[16] bookdown_0.3         caTools_1.17.1       scales_0.4.1         DEoptimR_1.0-8       mvtnorm_1.0-5       
[21] robustbase_0.92-7    stringr_1.1.0        digest_0.6.12        rmarkdown_1.3        wdman_0.2.2         
[26] rrcov_1.4-3          htmltools_0.3.5      WriteXLS_4.0.0       limma_3.28.21        RSQLite_1.1-2       
[31] shiny_1.0.0          gtools_3.5.0         dplyr_0.5.0          RCurl_1.95-4.8       magrittr_1.5        
[36] Rcpp_0.12.9          munsell_0.4.3        S4Vectors_0.10.3     viridis_0.3.4        stringi_1.1.2       
[41] yaml_2.1.14          edgeR_3.14.0         zlibbioc_1.18.0      rhdf5_2.16.0         gplots_3.0.1        
[46] Rtsne_0.11           plyr_1.8.4           grid_3.3.0           gdata_2.17.0         shinydashboard_0.5.3
[51] semver_0.2.0         lattice_0.20-34      knitr_1.15.1         rjson_0.2.15         rngtools_1.2.4      
[56] reshape2_1.4.2       codetools_0.2-15     biomaRt_2.28.0       stats4_3.3.0         XML_3.98-1.5        
[61] evaluate_0.10        data.table_1.10.4    httpuv_1.3.3         foreach_1.4.3        gtable_0.2.0        
[66] openssl_0.9.6        assertthat_0.1       binman_0.1.0         mime_0.5             xtable_1.8-2        
[71] e1071_1.6-8          rsconnect_0.7        class_7.3-14         pcaPP_1.9-61         tibble_1.2          
[76] pheatmap_1.0.8       iterators_1.0.8      RSelenium_1.7.1      AnnotationDbi_1.34.4 registry_0.3        
[81] memoise_1.0.0        IRanges_2.6.1        tximport_1.0.3       cluster_2.0.5        ROCR_1.0-7

sc3 scater single-cell single cell • 1.9k views

ADD COMMENT • link updated 7.2 years ago by Vladimir Kiselev ▴ 150 • written 7.2 years ago by Assa Yeroslaviz ★ 1.5k

0

Entering edit mode

I have updated the SC3 package to 1.3.14.

Now I get a different error message, which is even before the sc3_prepare error.

when running this command

patient.sceset <- sc3(patient.sceset, ks = 2:5, k_estimator = TRUE, biology = TRUE)

I get the following error

Setting SC3 parameters...
Error in (function (cl, name, valueClass)  :
  ‘sc3’ is not a slot in class “SCESet”

What is now wrong?

thanks again

Assa

P.S.

but at least I can now see the sc3_prepare function, so it probably exists, when I get there.

ADD REPLY • link 7.2 years ago Assa Yeroslaviz ★ 1.5k

0

Entering edit mode

Have you updated the scater package as well? You need the latest scater version (1.2.0) for SC3 to work.

ADD REPLY • link 7.2 years ago Vladimir Kiselev ▴ 150

0

Entering edit mode

thanks I have updated R to 3.3.2 and bio to 3.4. now it works. I do get a different error message, but this seems to be something else.

when running the jackStraw function it throws back an error -

> d <- jackStraw(
+     d,
+     num.replicate = 200,
+     do.print = FALSE,
+     num.pc = 25
+ )
Error in JackStraw(...) : Too few variable genes
> d
An object of class seurat in project NBT
 4260 genes across 51 samples.

I do need to take a closer look at Seurat though.

ADD REPLY • link 7.2 years ago Assa Yeroslaviz ★ 1.5k

0

Entering edit mode

Note that in the course we used an old version of Seurat (1.3). Now there is a new one available, where the clustering algorithm is completely different. We will try to update the course with this new version when we run it next time (middle of March 2017).

ADD REPLY • link 7.2 years ago Vladimir Kiselev ▴ 150

score 1 · Answer 1 · 2017-02-16

1

Entering edit mode

Vladimir Kiselev ▴ 150

@vladimir-kiselev-9342

Last seen 5.1 years ago

Sanger Institute, Cambridge, UK

Hi,

It looks like you have an old version of Bioconductor (both SC3 and scater versions are quite old). Could you please provide an output of this call?

source("https://bioconductor.org/biocLite.R")

If it offers you to upgrade, please do so and then update all the Bioconductor packages. This should solve your problem.

Cheers,

Vlad

ADD COMMENT • link 7.2 years ago Vladimir Kiselev ▴ 150