kharchenkolab / leidenalg Goto Github PK

How about making it broader and including all communities from Conos here? I'm a particularly big fan of the rleiden community.

Hi!
I have run leidenAlg::leiden.community() on the exact same graph g and with identical seeds on Windows and on Linux and the results differ. Is there a known reason (and maybe even fix) for this?

Cheers,
Marie

Code:
set.seed(168575)
partition <- leidenAlg::leiden.community(graph = g, n.iterations = 50)

Expected behaviour: partition is the same when running code on Windows and Linux.
Observed behaviour: partition is different.

hi,
Sorry to interrupt you,but i ran into a issue when i run this code in R:
mini_visium <- doLeidenCluster(gobject = mini_visium,
resolution = 0.4, n_iterations = 1000)

a problem occurs :
Error in py_run_file_impl(file, local, convert) :
ModuleNotFoundError: No module named 'leidenalg'

Detailed traceback:
File "", line 10, in
File "D:\Win10 System\Documents\R\win-library\4.1\reticulate\python\rpytools\loader.py", line 44, in _import_hook
level=level

So,i checked this :

And using conda, i found this: leidenalg is there!!!

And then i tried this : Although leidenalg , pandas AND pillow is in my evrironment,but only pandas could be uesed.

Could you please help sovle this problem?

Hi @evanbiederstedt ,

I'm trying to install leidenAlg on R-4.1.2 (libxml2, glpk and gmp are installed). Compilation goes fine, but the last step returns

** testing if installed package can be loaded from temporary location
Error: package or namespace load failed for ‘leidenAlg’ in dyn.load(file, DLLpath = DLLpath, ...):
 unable to load shared object '/home/vpetukhov/R/x86_64-pc-linux-gnu-library/4.1/00LOCK-leidenAlg/00new/leidenAlg/libs/leidenAlg.so':
  /home/vpetukhov/R/x86_64-pc-linux-gnu-library/4.1/00LOCK-leidenAlg/00new/leidenAlg/libs/leidenAlg.so: undefined symbol: igraph_rngtype_mt19937

The same result goes on my local laptop with Gentoo and on a server with RedHat 8. Do you have any idea how to debug this?

If I run devtools::load_all(), I can check the compilation files. Then, nm -D /tmp/Rtmpxafa69/pkgload32db541a1f34/leidenAlg.so shows that there is indeed a symbol U igraph_rngtype_mt19937.
And output of ldd /tmp/Rtmpxafa69/pkgload32db541a1f34/leidenAlg.so is the following:

	linux-vdso.so.1 (0x00007ffcac358000)
	libstdc++.so.6 => /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/libstdc++.so.6 (0x00007f9fa16b8000)
	libm.so.6 => /lib64/libm.so.6 (0x00007f9fa15e4000)
	libR.so => /usr/lib64/R/lib/libR.so (0x00007f9fa10d1000)
	libgcc_s.so.1 => /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/libgcc_s.so.1 (0x00007f9fa10b6000)
	libc.so.6 => /lib64/libc.so.6 (0x00007f9fa0ec3000)
	/lib64/ld-linux-x86-64.so.2 (0x00007f9fa1933000)
	libblas.so.3 => /usr/lib64/libblas.so.3 (0x00007f9fa0e1e000)
	libreadline.so.8 => /lib64/libreadline.so.8 (0x00007f9fa0dc3000)
	libpcre2-8.so.0 => /usr/lib64/libpcre2-8.so.0 (0x00007f9fa0d26000)
	liblzma.so.5 => /lib64/liblzma.so.5 (0x00007f9fa0cfb000)
	libbz2.so.1 => /lib64/libbz2.so.1 (0x00007f9fa0ce2000)
	libz.so.1 => /lib64/libz.so.1 (0x00007f9fa0cc4000)
	libtirpc.so.3 => /lib64/libtirpc.so.3 (0x00007f9fa0c9b000)
	libdl.so.2 => /lib64/libdl.so.2 (0x00007f9fa0c94000)
	libicuuc.so.69 => /usr/lib64/libicuuc.so.69 (0x00007f9fa0a86000)
	libicui18n.so.69 => /usr/lib64/libicui18n.so.69 (0x00007f9fa0744000)
	libgomp.so.1 => /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/libgomp.so.1 (0x00007f9fa0702000)
	libpthread.so.0 => /lib64/libpthread.so.0 (0x00007f9fa06fd000)
	libgfortran.so.5 => /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/libgfortran.so.5 (0x00007f9fa0441000)
	libtinfow.so.6 => /lib64/libtinfow.so.6 (0x00007f9fa03f6000)
	libicudata.so.69 => /usr/lib64/libicudata.so.69 (0x00007f9f9e89b000)
	libquadmath.so.0 => /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/libquadmath.so.0 (0x00007f9f9e851000)

I tried to randomly tweak Makevars with no success.

Hi,

I've been using the rleiden.community() function for clustering detection. However, I've noticed that when I run the same analysis multiple times, results tend to change. I also tried to set a seed, but even with the same seed results change.

Do you have any suggestion on how to prevent this behavior and make results reproducible?

Thank you.

Hi, hanks for this great package! I cant seem to install it on a Mac though. Here is the sessionInfo and the errors.

• Session info

sessionInfo()
R version 4.0.5 (2021-03-31)
Platform: x86_64-apple-darwin17.0 (64-bit)
Running under: macOS Catalina 10.15.7

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] stats graphics grDevices utils datasets methods base

other attached packages:
[1] igraph_1.2.6 Matrix_1.3-3

loaded via a namespace (and not attached):
[1] Rcpp_1.0.6 lattice_0.20-44 png_0.1-7 shinyWidgets_0.6.0 assertthat_0.2.1 rprojroot_2.0.2
[7] digest_0.6.27 utf8_1.2.1 mime_0.10 R6_2.5.0 plyr_1.8.6 evaluate_0.14
[13] ggplot2_3.3.3 pillar_1.6.0 rlang_0.4.11 sortable_0.4.4 jquerylib_0.1.4 learnr_0.10.1
[19] sccore_0.1.3 reticulate_1.20 grr_0.9.5 rmarkdown_2.8 stringr_1.4.0 htmlwidgets_1.5.3
[25] munsell_0.5.0 tinytex_0.31 shiny_1.6.0 compiler_4.0.5 httpuv_1.6.1 xfun_0.22
[31] pkgconfig_2.0.3 clipr_0.7.1 htmltools_0.5.1.1 tidyselect_1.1.1 tibble_3.1.1 gridExtra_2.3
[37] dendextend_1.15.1 fansi_0.4.2 viridisLite_0.4.0 withr_2.4.2 crayon_1.4.1 dplyr_1.0.6
[43] later_1.2.0 Matrix.utils_0.9.8 grid_4.0.5 jsonlite_1.7.2 xtable_1.8-4 gtable_0.3.0
[49] lifecycle_1.0.0 DBI_1.1.1 magrittr_2.0.1 scales_1.1.1 stringi_1.5.3 reshape2_1.4.4
[55] viridis_0.6.1 promises_1.2.0.1 bslib_0.2.4 ellipsis_0.3.2 generics_0.1.0 vctrs_0.3.8
[61] RColorBrewer_1.1-2 tools_4.0.5 glue_1.4.2 markdown_1.1 purrr_0.3.4 shinycssloaders_1.0.0
[67] parallel_4.0.5 fastmap_1.1.0 yaml_2.2.1 colorspace_2.0-1 knitr_1.33 sass_0.3.1

• Installing from cran works fine but package will not load:

library(leidenAlg)
Loading required package: Matrix
Loading required package: igraph

Attaching package: ‘igraph’

The following objects are masked from ‘package:stats’:

decompose, spectrum

The following object is masked from ‘package:base’:

union

Error: package or namespace load failed for ‘leidenAlg’ in dyn.load(file, DLLpath = DLLpath, ...):
unable to load shared object '/Library/Frameworks/R.framework/Versions/4.0/Resources/library/leidenAlg/libs/leidenAlg.so':
dlopen(/Library/Frameworks/R.framework/Versions/4.0/Resources/library/leidenAlg/libs/leidenAlg.so, 6): Library not loaded: @rpath/igraph.so
Referenced from: /Library/Frameworks/R.framework/Versions/4.0/Resources/library/leidenAlg/libs/leidenAlg.so
Reason: image not found

• Installing from github crashes before even loading. This is not the full console output, i think this is the relevant parts

devtools::install_github('kharchenkolab/leidenAlg', build_vignettes = TRUE)
Downloading GitHub repo kharchenkolab/leidenAlg@HEAD
✓ checking for file ‘/private/var/folders/sy/gp7_zd0918j429vrk56lxvgw0000gn/T/RtmpOElSu6/remotes5c113b358a16/kharchenkolab-leidenAlg-0db1cdf/DESCRIPTION’ ...

• Some warnings, this is the last messages

Attaching package: ‘igraph’

The following objects are masked from ‘package:stats’:

decompose, spectrum

The following object is masked from ‘package:base’:

union

clang++ -mmacosx-version-min=10.13 -std=gnu++11 -dynamiclib -Wl,-headerpad_max_install_names -undefined dynamic_lookup -single_module -multiply_defined suppress -L/Library/Frameworks/R.framework/Resources/lib -L/usr/local/opt/gettext/lib -L/usr/local/opt/llvm/lib -o leidenAlg.so RcppExports.o leiden.o -L/usr/lib/ -L. -lpthread -lstdc++ -lleidenalg -lm -L/Library/Frameworks/R.framework/Resources/lib -lRlapack -L/Library/Frameworks/R.framework/Resources/lib -lRblas -L/usr/local/gfortran/lib/gcc/x86_64-apple-darwin18/8.2.0 -L/usr/local/gfortran/lib -lgfortran -lquadmath -lm /Library/Frameworks/R.framework/Versions/4.0/Resources/library/igraph/libs/igraph.so -F/Library/Frameworks/R.framework/.. -framework R -Wl,-framework -Wl,CoreFoundation

Attaching package: ‘igraph’

The following objects are masked from ‘package:stats’:

decompose, spectrum

The following object is masked from ‘package:base’:

union

ld: warning: directory not found for option '-L/usr/local/opt/gettext/lib'
ld: warning: directory not found for option '-L/usr/local/opt/llvm/lib'
ld: warning: directory not found for option '-L/usr/local/gfortran/lib/gcc/x86_64-apple-darwin18/8.2.0'
ld: warning: directory not found for option '-L/usr/local/gfortran/lib'
ld: library not found for -lgfortran
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make: *** [leidenAlg.so] Error 1
ERROR: compilation failed for package ‘leidenAlg’

• removing ‘/Library/Frameworks/R.framework/Versions/4.0/Resources/library/leidenAlg’
• restoring previous ‘/Library/Frameworks/R.framework/Versions/4.0/Resources/library/leidenAlg’
  Warning message:
  In i.p(...) :
  installation of package ‘/var/folders/sy/gp7_zd0918j429vrk56lxvgw0000gn/T//RtmpOElSu6/file5c1134ddc4bf/leidenAlg_0.1.1.tar.gz’ had non-zero exit status

Thanks for any and all help!
Best

Hi, I am having issues installing leidenAlg in a conda environment on macOS. The installation is looking for an igraph.so file where only igraph.dylib is available, since I'm on macOS.

I also looked at #5, but all mentioned dependencies are installed within the conda environment.

Can you maybe help out?

Support is much appreciated.

• Error message:

> install.packages('leidenAlg')

...

Attaching package: ‘igraph’

The following objects are masked from ‘package:stats’:

    decompose, spectrum

The following object is masked from ‘package:base’:

    union

clang-12: error: no such file or directory: '/.../miniconda3/envs/scib-R/lib/R/library/igraph/libs/igraph.so'
make: *** [/.../miniconda3/envs/scib-R/lib/R/share/make/shlib.mk:6: leidenAlg.dylib] Error 1
ERROR: compilation failed for package ‘leidenAlg’
* removing ‘/.../miniconda3/envs/scib-R/lib/R/library/leidenAlg’

• SessionInfo. I'm limited to R 3.6 due to another package.

> sessionInfo()
R version 3.6.0 (2019-04-26)
Platform: x86_64-apple-darwin13.4.0 (64-bit)
Running under: macOS High Sierra 10.13.6

Matrix products: default
BLAS/LAPACK: /.../miniconda3/envs/scib-R/lib/R/lib/libRblas.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] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] igraph_1.2.9

I'm experimenting the leidenalg community detection and have noticed that weights is not affecting the results? also i have inverted weights to double check and still, weights not doing a noticeable impact on the results. my cods:

optimiser = la.Optimiser()
optimiser.set_rng_seed(0)
partitions = la.ModularityVertexPartition(g, weights='weights')

diff = 1
_ = datetime.datetime.now().timestamp()

while diff > 0 and (datetime.datetime.now().timestamp() - _) < 30:
diff = optimiser.optimise_partition(partitions)

My question is: am I missing something?

Dear team,
I am trying to run infercnv which runs well till it gets to the step when it needs leidenalg. I installed leidenalg via pip install and it is installed but I consistently get an ImportError that leidenalg module is not found. So , I installed leindenAlg and loaded it successfully in R but I am still getting the same error that R can't see the package?.

Any suggestions from your side, please?.

I have asked leidenalg team and infercnv as well

Thanks

We're changing the layout of igraph objects. This leads to check failures in your package, see https://github.com/igraph/rigraph/blob/f-igraph-t-idx-revdepcheck/revdep/problems.md and igraph/rigraph#789 for details.

To reproduce, please install the development version of igraph via

# install.packages("pak")
pak::pak("igraph/rigraph")

and run R CMD check on your package.

We plan to release an igraph update on June 12, two weeks from now. Can you please send an update to CRAN that fixes the checks?

This package seems to use its own copy of igraph. The error happens in R_SEXP_to_igraph() which still assumes the old format. Please retrieve an edgelist on the R side and construct your igraph object on the C/C++ side.

Happy to help.

Hi

I have some problems loading the package. The following error message appears:

`The following object is masked from ‘package:base’:

    union

Error: package or namespace load failed for ‘leidenAlg’ in dyn.load(file, DLLpath = DLLpath, ...):
 unable to load shared object '/Library/Frameworks/R.framework/Versions/3.6/Resources/library/leidenAlg/libs/leidenAlg.so':
  dlopen(/Library/Frameworks/R.framework/Versions/3.6/Resources/library/leidenAlg/libs/leidenAlg.so, 6): Library not loaded: @rpath/igraph.so
  Referenced from: /Library/Frameworks/R.framework/Versions/3.6/Resources/library/leidenAlg/libs/leidenAlg.so
  Reason: image not found`

R. Version()

platform       x86_64-apple-darwin15.6.0   
arch           x86_64                      
os             darwin15.6.0                
system         x86_64, darwin15.6.0        
status                                     
major          3                           
minor          6.2                         
year           2019                        
month          12                          
day            12                          
svn rev        77560                       
language       R                           
version.string R version 3.6.2 (2019-12-12)
nickname       Dark and Stormy Night 

Any tips?

In the leiden.cpp source code, can we add another level of for loop to run it multiple times with the random number being updated? and after the while loop ends, access the quality or modularity so we can finally pick the best result. This is something that Seurat's louvain implementation has and I figured this should also be helpful for running Leiden.

Thanks,
Yichen

When supplying an unweighted graph (for example, the output of igraph::graph_from_adj_list) I get:

Error in find_partition(graph, igraph::E(graph)$weight, resolution, n.iterations) :
Not compatible with requested type: [type=NULL; target=double].

This is fixed if I manually set all weights to 1. It would be nice if this were the default behavior for unweighted graphs. Otherwise great implementation!