Open nick-youngblut opened 3 years ago
I tried a conda env with liblapack 3.9.0 7_openblas conda-forge
installed versus liblapack 3.9.0 9_mkl conda-forge
. The openblas version works with larger distance matrices.
Installing liblapack>=3.9.0=7_openblas
over the 9_mkl
version in my other conda env required upgrading/changing basically ever package installed, but that fixed the issue
Installing liblapack>=3.9.0=7_openblas
then breaks bioconductor-deseq2. Specifically, the dependency GenomeInfoDb
then cannot be installed. If I uninstall bioconductor-deseq2
or bioconductor-genomeinfodbdata
, conda tried to re-install liblapack 3.9.0 9_mkl conda-forge
!!!
If I use conda install bioconductor-deseq2 "liblapack>=3.9.0=7_openblas"
, to re-install the bioconductor packages, I just get the same error as before:
Loading required package: GenomeInfoDb
Error: package or namespace load failed for ‘GenomeInfoDb’ in loadNamespace(i, c(lib.loc, .libPaths()), versionCheck = vI[[i]]):
there is no package called ‘GenomeInfoDbData’
Error: package ‘GenomeInfoDb’ could not be loaded
Traceback:
1. library(DESeq2)
2. .getRequiredPackages2(pkgInfo, quietly = quietly)
3. library(pkg, character.only = TRUE, logical.return = TRUE, lib.loc = lib.loc,
. quietly = quietly)
4. .getRequiredPackages2(pkgInfo, quietly = quietly)
5. stop(gettextf("package %s could not be loaded", sQuote(pkg)),
. call. = FALSE, domain = NA)
I'm stuck in dependency conflict hell. This seems to happen quite a bit with conda & R.
@isuruf, could you transfer this to the intel_repack feedstock? It sounds like an MKL issue to me.
I am having a similar issue when trrying to generate a binary matrix from a large dataset. I tried everything, do you have a solution? Thanks! G.
library(phyloseq)
> pcoa_ecm_jac <- ordinate(physeq_ecm, method = "PCoA", distance = "jaccard", binary=TRUE)
Error in eigen(delta1) : error code 1 from Lapack routine 'dsyevr'
> sessionInfo()
R version 4.3.1 (2023-06-16)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 22.04.2 LTS
Matrix products: default
BLAS/LAPACK: /usr/lib/x86_64-linux-gnu/libmkl_rt.so; LAPACK version 3.8.0
locale:
[1] LC_CTYPE=en_CA.UTF-8 LC_NUMERIC=C LC_TIME=en_CA.UTF-8 LC_COLLATE=en_CA.UTF-8 LC_MONETARY=en_CA.UTF-8 LC_MESSAGES=en_CA.UTF-8
[7] LC_PAPER=en_CA.UTF-8 LC_NAME=C LC_ADDRESS=C LC_TELEPHONE=C LC_MEASUREMENT=en_CA.UTF-8 LC_IDENTIFICATION=C
time zone: America/Toronto
tzcode source: system (glibc)
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] lubridate_1.9.2 forcats_1.0.0 stringr_1.5.0 dplyr_1.1.2 purrr_1.0.1 readr_2.1.4 tidyr_1.3.0
[8] tibble_3.2.1 ggplot2_3.4.2 tidyverse_2.0.0 phyloseq_1.44.0 vegan_2.6-4 lattice_0.21-8 permute_0.9-7
[15] BiocManager_1.30.21.1
loaded via a namespace (and not attached):
[1] gtable_0.3.3 rhdf5_2.44.0 Biobase_2.60.0 tzdb_0.4.0 generics_0.1.3 rhdf5filters_1.12.1
[7] vctrs_0.6.3 tools_4.3.1 bitops_1.0-7 biomformat_1.28.0 stats4_4.3.1 parallel_4.3.1
[13] fansi_1.0.4 cluster_2.1.4 pkgconfig_2.0.3 Matrix_1.6-0 data.table_1.14.8 S4Vectors_0.38.1
[19] lifecycle_1.0.3 GenomeInfoDbData_1.2.10 compiler_4.3.1 Biostrings_2.68.1 munsell_0.5.0 codetools_0.2-19
[25] GenomeInfoDb_1.36.1 RCurl_1.98-1.12 pillar_1.9.0 crayon_1.5.2 MASS_7.3-60 iterators_1.0.14
[31] foreach_1.5.2 nlme_3.1-162 tidyselect_1.2.0 digest_0.6.33 stringi_1.7.12 reshape2_1.4.4
[37] splines_4.3.1 ade4_1.7-22 grid_4.3.1 colorspace_2.1-0 cli_3.6.1 magrittr_2.0.3
[43] survival_3.5-5 utf8_1.2.3 ape_5.7-1 withr_2.5.0 scales_1.2.1 timechange_0.2.0
[49] XVector_0.40.0 igraph_1.5.0.1 multtest_2.56.0 hms_1.1.3 IRanges_2.34.1 mgcv_1.9-0
[55] rlang_1.1.1 Rcpp_1.0.11 glue_1.6.2 BiocGenerics_0.46.0 rstudioapi_0.15.0 jsonlite_1.8.7
[61] R6_2.5.1 Rhdf5lib_1.22.0 plyr_1.8.8 zlibbioc_1.46.0
Does anyone here found a solution?
I recently updated my conda env (see below), which broke the R function
cmdscale()
for for larger distance matrices. If I provide a distance matrix larger than the lower triangle of a 200 x 200 matrix, I get the error:The function works with any "small" distance matrix.
My conda env:
conda info: