Closed Acribbs closed 4 years ago
Great, do you want to wait, I’m refactoring this pipeline to change the various inputs and remove sailfish?
ok, if your refactoring this pipeline then we can wait to merge. Will set another branch called testing for making changes to testing pipeline
Hi, can we add r-wasabi to the current conda build? It is a requirement for the rnaseqdiffexperssion pipeline. As part of removing some of the duplication of the code I decided to run the salmon quantifier from rnaseq.py in the ranseqqc pipeline. This gets around the problem that sailfish used to generate both transcript and gene counts whereas salmon no longer offers transcript2genemap conversions and therefore only outputs transcript counts (gene count instead done by tximport directly in R now). This is all for good reasons, however for the rnaseqqc pipeline depends on gene counts. The downside of using the rnaseq.py code is that this behemoth created by Tom many moons ago, allows for conversion between all types of transcript/gene abundance calculators and all types of differenential expression tools and therefore requires the conversion of salmon to kallisto format by default. R Wasabi is required for that.
Ah I hadn’t realised and it makes perfect sense to use r-wasabi. Just add it to the conda requirements in rnaseqdiffexpression.
Would we need a complete refactoring of rnaseqdiffexpressiin? Happy to help as I use this pipeline quite a lot
It's already in rnaseqdiffexpression (and in the pipeline_rnaseqdiffexpression.yml, where it is actually hashed out). I would now need to add it to pipeline_rnaseqqc.yml?
I did try to install it and it completely changes the conda environment and results in 50-100 package updates and installs
Given that it's an integral part of running salmon in rnaseqdiffexpression.py and the pipeline fails without r-wasabi, how does travis get through ranseqdiffexpression tests (is salmon run in the tests?)?
Hi,
I would like to have a look at this with more detail after work.
Travis do not run any pipeline tests. Those are performed by Jenkins, which I am troubleshooting at the moment.
I will add https://anaconda.org/bioconda/r-wasabi to the conda env asap.
Best regards, Sebastian
PS: As per your environment, please share what command you use to install it and the conda output.
Yes, apologies I meant Jenkins, I still misremember which is which.
conda install -c bioconda r-wasabi
I did not install it because of the many conflicts.
Did you install your cgat environment with the installer?
Could you please share the output of conda info
?
active environment : cgat-flow
active env location : /XXX/cgat-developers-v2/conda-install/envs/cgat-flow
shell level : 2
user config file : ~/.condarc
populated config files : ~/.condarc
conda version : 4.7.12
conda-build version : not installed
python version : 3.7.3.final.0
virtual packages :
base environment : /XXX/conda-install (writable)
channel URLs : https://conda.anaconda.org/conda-forge/linux-64
https://conda.anaconda.org/conda-forge/noarch
https://conda.anaconda.org/bioconda/linux-64
https://conda.anaconda.org/bioconda/noarch
https://conda.anaconda.org/cgat/linux-64
https://conda.anaconda.org/cgat/noarch
https://repo.anaconda.com/pkgs/main/linux-64
https://repo.anaconda.com/pkgs/main/noarch
https://repo.anaconda.com/pkgs/r/linux-64
https://repo.anaconda.com/pkgs/r/noarch
package cache : /XXX/conda-install/pkgs
~/.conda/pkgs
envs directories : /XXX/conda-install/envs
~/.conda/envs
platform : linux-64
user-agent : conda/4.7.12 requests/2.22.0 CPython/3.7.3 Linux/2.6.32-754.el6.x86_64 scientific/6.10 glibc/2.12
UID:GID : 1400:1000
netrc file : None
offline mode : False
Recent re-install, I do believe I made a minor change by updating libboost on cgat-flow but otherwise no changes to conda from the original install.
Not sure you should have:
https://conda.anaconda.org/cgat/linux-64
https://conda.anaconda.org/cgat/noarch
in your .condarc
file. Could you please share cat ~/.condarc
Anyway, could you please try instead:
conda install --dry-run -c conda-forge -c bioconda -c defaults r-wasabi
and share the output?
Sorry, forgot a key flag --no-update-deps
:
conda install --dry-run --no-update-deps -c conda-forge -c bioconda -c defaults r-wasabi
r-wasabi has very few dependancies, however it depends on bioconductor-rhdf5 and in the past I have had a lot of problems with installing conda packages that depend on this (Think it was the tximport or the one of the packages for Deseq2, I cant remember which ones now though).
and share the output?
- conda-forge - https://conda.anaconda.org/bioconda - https://conda.anaconda.org/cgat - defaults
Thanks, please also share the output of:
conda install --dry-run --no-update-deps -c conda-forge -c bioconda -c defaults r-wasabi
Thanks, please also share the output of:
conda install --dry-run --no-update-deps -c conda-forge -c bioconda -c defaults r-wasabi
Yes, coming, it's still solving... takes about 20 minutes.
In the meantime, I suggest you have update your ~/.condarc
with:
- conda-forge
- bioconda
- defaults
Also, if you have a recent re-install, could you please share conda list rpy2
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Collecting package metadata (repodata.json): done
Solving environment: done
## Package Plan ##
environment location: /ifs/projects/jakubs/cgat-developers-v2/conda-install/envs/cgat-flow
added / updated specs:
- r-wasabi
The following packages will be downloaded:
package | build
---------------------------|-----------------
bioconductor-annotate-1.60.1| r351_0 2.1 MB bioconda
bioconductor-annotationdbi-1.44.0| r351_0 4.9 MB bioconda
bioconductor-biobase-2.42.0| r351h14c3975_1 2.3 MB bioconda
bioconductor-biocgenerics-0.28.0| r351_1 695 KB bioconda
bioconductor-biocparallel-1.16.6| r351h1c2f66e_0 1.1 MB bioconda
bioconductor-biomart-2.38.0| r351_0 526 KB bioconda
bioconductor-biostrings-2.50.2| r351h14c3975_0 13.8 MB bioconda
bioconductor-biovizbase-1.18.0| 1 2.2 MB bioconda
bioconductor-bsgenome-1.50.0| r351_0 6.9 MB bioconda
bioconductor-cummerbund-2.12.1| r351_1 2.8 MB bioconda
bioconductor-delayedarray-0.8.0| r351h14c3975_0 1.6 MB bioconda
bioconductor-deseq-1.34.1 | r351h14c3975_0 1.8 MB bioconda
bioconductor-deseq2-1.22.1 | r351hf484d3e_0 2.5 MB bioconda
bioconductor-edger-3.26.0 | r351hf484d3e_0 2.2 MB bioconda
bioconductor-genefilter-1.64.0| r351h1c2f66e_1 1.7 MB bioconda
bioconductor-geneplotter-1.60.0| r351_0 1.5 MB bioconda
bioconductor-genomeinfodb-1.18.1| r351_0 3.8 MB bioconda
bioconductor-genomeinfodbdata-1.2.1| r351_0 6 KB bioconda
bioconductor-genomicalignments-1.18.1| r351h14c3975_0 2.2 MB bioconda
bioconductor-genomicfeatures-1.34.1| r351_0 2.1 MB bioconda
bioconductor-genomicranges-1.34.0| r351h14c3975_0 2.1 MB bioconda
bioconductor-gviz-1.14.2 | 0 4.0 MB bioconda
bioconductor-hpar-1.26.0 | r351_0 7.2 MB bioconda
bioconductor-iranges-2.16.0| r351h14c3975_0 2.3 MB bioconda
bioconductor-limma-3.40.0 | r351h14c3975_0 2.6 MB bioconda
bioconductor-rhdf5-2.28.0 | r351hf484d3e_0 2.4 MB bioconda
bioconductor-rhdf5lib-1.6.0| r351h14c3975_0 3.6 MB bioconda
bioconductor-rsamtools-1.34.0| r351hf484d3e_0 3.8 MB bioconda
bioconductor-rtracklayer-1.42.1| r351h9d9f1b6_1 2.7 MB bioconda
bioconductor-s4vectors-0.20.1| r351h14c3975_0 2.0 MB bioconda
bioconductor-summarizedexperiment-1.12.0| r351_0 2.6 MB bioconda
bioconductor-variantannotation-1.28.3| r351h14c3975_0 3.3 MB bioconda
bioconductor-xvector-0.22.0| r351h14c3975_0 711 KB bioconda
bioconductor-zlibbioc-1.28.0| r351h14c3975_0 112 KB bioconda
certifi-2019.9.11 | py36_0 147 KB conda-forge
cloog-0.18.0 | 0 565 KB
gcc-4.8.3 | 1 66.5 MB cgat
gmp-6.1.2 | hf484d3e_1000 751 KB conda-forge
isl-0.21 | he80fd80_0 1.5 MB conda-forge
mpc-1.1.0 | h04dde30_1006 149 KB conda-forge
mpfr-4.0.2 | he80fd80_0 673 KB conda-forge
openssl-1.1.1d | h516909a_0 2.1 MB conda-forge
r-3.1.2 | 2 18.5 MB
r-acepack-1.4.1 | r35h9bbef5b_1004 69 KB conda-forge
r-assertthat-0.2.1 | r35h6115d3f_1 69 KB conda-forge
r-backports-1.1.5 | r35hcdcec82_0 67 KB conda-forge
r-base-3.5.1 | h08e1455_1008 39.2 MB conda-forge
r-base64enc-0.1_3 | r35hcdcec82_1003 43 KB conda-forge
r-bh-1.69.0_1 | r35h6115d3f_1 10.7 MB conda-forge
r-bit-1.1_14 | r35hcdcec82_1 243 KB conda-forge
r-bit64-0.9_7 | r35hcdcec82_1001 477 KB conda-forge
r-bitops-1.0_6 | r35hcdcec82_1003 39 KB conda-forge
r-blob-1.1.1 | r351_1001 27 KB conda-forge
r-catools-1.17.1.2 | r35h0357c0b_1 223 KB conda-forge
r-checkmate-1.9.4 | r35hcdcec82_1 647 KB conda-forge
r-cli-1.1.0 | r35h6115d3f_2 177 KB conda-forge
r-cluster-2.1.0 | r35h9bbef5b_2 551 KB conda-forge
r-colorspace-1.4_1 | r35hcdcec82_1 2.5 MB conda-forge
r-crayon-1.3.4 | r35h6115d3f_1002 745 KB conda-forge
r-crosstalk-1.0.0 | r35h6115d3f_1002 596 KB conda-forge
r-curl-4.2 | r35hcdcec82_0 697 KB conda-forge
r-data.table-1.12.6 | r35hcdcec82_0 1.8 MB conda-forge
r-dbi-1.0.0 | r35h6115d3f_1002 883 KB conda-forge
r-dichromat-2.0_0 | r35_2001 151 KB conda-forge
r-digest-0.6.22 | r35h0357c0b_0 192 KB conda-forge
r-dt-0.9 | r35h6115d3f_0 888 KB conda-forge
r-evaluate-0.14 | r35h6115d3f_1 79 KB conda-forge
r-fansi-0.4.0 | r35hcdcec82_1001 192 KB conda-forge
r-fastcluster-1.1.25 | r35hc99cbb1_1002 191 KB conda-forge
r-foreign-0.8_72 | r35hcdcec82_0 266 KB conda-forge
r-formatr-1.7 | r35h6115d3f_1 165 KB conda-forge
r-formula-1.2_3 | r35h6115d3f_1002 192 KB conda-forge
r-futile.logger-1.4.3 | r35h6115d3f_1002 107 KB conda-forge
r-futile.options-1.0.1 | r35h6115d3f_1001 23 KB conda-forge
r-gdata-2.18.0 | r35h6115d3f_1002 1.1 MB conda-forge
r-ggplot2-2.2.1 | r351h6115d3f_1 3.1 MB conda-forge
r-glue-1.3.1 | r35hcdcec82_1 165 KB conda-forge
r-gmd-0.3.3 | r35h516909a_1002 505 KB conda-forge
r-gplots-3.0.1.1 | r35h6115d3f_1 656 KB conda-forge
r-gridextra-2.3 | r35h6115d3f_1002 1.1 MB conda-forge
r-gsalib-2.1 | r35_1001 29 KB conda-forge
r-gtable-0.3.0 | r35h6115d3f_2 420 KB conda-forge
r-gtools-3.8.1 | r35hcdcec82_1003 324 KB conda-forge
r-highr-0.8 | r35h6115d3f_1 47 KB conda-forge
r-hmisc-4.2_0 | r35h9bbef5b_2 3.0 MB conda-forge
r-hms-0.4.2 |r351h6115d3f_1000 71 KB conda-forge
r-htmltable-1.13.2 | r35h6115d3f_0 328 KB conda-forge
r-htmltools-0.4.0 | r35h0357c0b_0 213 KB conda-forge
r-htmlwidgets-1.5.1 | r35h6115d3f_0 674 KB conda-forge
r-httpuv-1.5.2 | r35h0357c0b_1 855 KB conda-forge
r-httr-1.4.1 | r35h6115d3f_1 485 KB conda-forge
r-jsonlite-1.6 | r35hcdcec82_1001 1006 KB conda-forge
r-kernsmooth-2.23_15 | r35h9bbef5b_1004 102 KB conda-forge
r-knitr-1.25 | r35h6115d3f_0 1.3 MB conda-forge
r-labeling-0.3 | r35h6115d3f_1002 65 KB conda-forge
r-lambda.r-1.2.4 | r35h6115d3f_0 118 KB conda-forge
r-later-1.0.0 | r35h0357c0b_0 140 KB conda-forge
r-lattice-0.20_38 | r35hcdcec82_1002 1.1 MB conda-forge
r-latticeextra-0.6_28 | r35h6115d3f_1002 2.1 MB conda-forge
r-lazyeval-0.2.2 | r35hcdcec82_1 163 KB conda-forge
r-locfit-1.5_9.1 | r35h516909a_1004 553 KB conda-forge
r-magrittr-1.5 | r35h6115d3f_1002 165 KB conda-forge
r-markdown-1.1 | r35hcdcec82_0 143 KB conda-forge
r-mass-7.3_51.4 | r35hcdcec82_1 1.1 MB conda-forge
r-matrix-1.2_17 | r35hcdcec82_1 3.8 MB conda-forge
r-matrixstats-0.55.0 | r35hcdcec82_0 904 KB conda-forge
r-memoise-1.1.0 | r35h6115d3f_1003 41 KB conda-forge
r-mime-0.7 | r35hcdcec82_1 51 KB conda-forge
r-munsell-0.5.0 | r35h6115d3f_1002 244 KB conda-forge
r-nnet-7.3_12 | r35hcdcec82_1003 130 KB conda-forge
r-openssl-1.1 |r351h3a9d887_1002 1.1 MB conda-forge
r-pillar-1.3.0 | r351h6115d3f_0 152 KB
r-pkgconfig-2.0.3 | r35h6115d3f_0 24 KB conda-forge
r-plogr-0.2.0 | r35h6115d3f_1002 18 KB conda-forge
r-plotrix-3.7_6 | r35h6115d3f_1 1.1 MB conda-forge
r-plyr-1.8.4 | r35h0357c0b_1003 814 KB conda-forge
r-prettyunits-1.0.2 | r35h6115d3f_1002 35 KB conda-forge
r-progress-1.2.2 | r35h6115d3f_1 87 KB conda-forge
r-promises-1.1.0 | r35h0357c0b_0 1.2 MB conda-forge
r-r6-2.4.0 | r35h6115d3f_2 62 KB conda-forge
r-rcolorbrewer-1.1_2 | r35h6115d3f_1002 59 KB conda-forge
r-rcpp-1.0.3 | r35h0357c0b_0 1.9 MB conda-forge
r-rcpparmadillo-0.9.800.1.0| r35h0357c0b_0 1.2 MB conda-forge
r-rcurl-1.95_4.12 | r35hcdcec82_1 912 KB conda-forge
r-reshape-0.8.8 | r35hcdcec82_1 173 KB conda-forge
r-reshape2-1.4.3 | r35h0357c0b_1004 129 KB conda-forge
r-rjson-0.2.20 | r35h0357c0b_1001 145 KB conda-forge
r-rlang-0.4.1 | r35hcdcec82_0 1021 KB conda-forge
r-rmarkdown-1.16 | r35h6115d3f_0 3.0 MB conda-forge
r-rpart-4.1_15 | r35hcdcec82_1 737 KB conda-forge
r-rsqlite-2.1.2 | r35h0357c0b_1 1.0 MB conda-forge
r-rstudioapi-0.10 | r35h6115d3f_2 217 KB conda-forge
r-scales-1.0.0 | r35h0357c0b_1002 576 KB conda-forge
r-shiny-1.3.2 | r35h6115d3f_1 4.2 MB conda-forge
r-snow-0.4_3 | r35h6115d3f_1001 122 KB conda-forge
r-sourcetools-0.1.7 | r35he1b5a44_1001 51 KB conda-forge
r-stringi-1.4.3 | r35h0357c0b_2 766 KB conda-forge
r-stringr-1.4.0 | r35h6115d3f_1 207 KB conda-forge
r-survival-2.44_1.1 | r35hcdcec82_1 4.8 MB conda-forge
r-tibble-1.4.2 |r351h96ca727_1002 229 KB conda-forge
r-tinytex-0.17 | r35h6115d3f_0 103 KB conda-forge
r-utf8-1.1.4 | r35hcdcec82_1001 158 KB conda-forge
r-venndiagram-1.6.20 | r35h6115d3f_1001 263 KB conda-forge
r-viridis-0.5.1 | r35h6115d3f_1003 1.8 MB conda-forge
r-viridislite-0.3.0 | r35h6115d3f_1002 63 KB conda-forge
r-wasabi-1.0.1 | r351_0 31 KB bioconda
r-xfun-0.10 | r35h6115d3f_0 182 KB conda-forge
r-xml-3.98_1.20 | r35hcdcec82_1 2.0 MB conda-forge
r-xtable-1.8_4 | r35h6115d3f_2 697 KB conda-forge
r-yaml-2.2.0 | r35hcdcec82_1002 113 KB conda-forge
rpy2-3.1.0 |py36r35hc1659b7_2 305 KB conda-forge
------------------------------------------------------------
Total: 306.4 MB
The following NEW packages will be INSTALLED:
bioconductor-rhdf5 bioconda/linux-64::bioconductor-rhdf5-2.28.0-r351hf484d3e_0
bioconductor-rhdf~ bioconda/linux-64::bioconductor-rhdf5lib-1.6.0-r351h14c3975_0
cloog pkgs/main/linux-64::cloog-0.18.0-0
gcc cgat/linux-64::gcc-4.8.3-1
gmp conda-forge/linux-64::gmp-6.1.2-hf484d3e_1000
isl conda-forge/linux-64::isl-0.21-he80fd80_0
mpc conda-forge/linux-64::mpc-1.1.0-h04dde30_1006
mpfr conda-forge/linux-64::mpfr-4.0.2-he80fd80_0
r pkgs/r/linux-64::r-3.1.2-2
r-rjson conda-forge/linux-64::r-rjson-0.2.20-r35h0357c0b_1001
r-wasabi bioconda/noarch::r-wasabi-1.0.1-r351_0
The following packages will be REMOVED:
bioconductor-annotationfilter-1.8.0-r36_1
bioconductor-ensembldb-2.8.0-r36_1
bioconductor-protgenerics-1.16.0-r36_1
bioconductor-rhtslib-1.16.1-r36hbcae180_1
r-askpass-1.1-r36hcdcec82_1
r-ellipsis-0.3.0-r36hcdcec82_0
r-fastmap-1.0.1-r36h0357c0b_0
r-mgcv-1.8_29-r36hcdcec82_0
r-nlme-3.1_141-r36h9bbef5b_1
r-sys-3.3-r36hcdcec82_0
r-vctrs-0.2.0-r36hcdcec82_1
r-withr-2.1.2-r36h6115d3f_1001
r-zeallot-0.1.0-r36h6115d3f_1001
The following packages will be UPDATED:
r-rcpp 1.0.2-r36h0357c0b_0 --> 1.0.3-r35h0357c0b_0
r-tinytex 0.16-r36h6115d3f_0 --> 0.17-r35h6115d3f_0
rpy2 3.1.0-py36r36hc1659b7_1 --> 3.1.0-py36r35hc1659b7_2
The following packages will be SUPERSEDED by a higher-priority channel:
bioconductor-cumm~ bioconda/noarch::bioconductor-cummerb~ --> bioconda/linux-64::bioconductor-cummerbund-2.12.1-r351_1
bioconductor-gviz bioconda/noarch::bioconductor-gviz-1.~ --> bioconda/linux-64::bioconductor-gviz-1.14.2-0
ca-certificates anaconda::ca-certificates-2019.10.16-0 --> conda-forge::ca-certificates-2019.9.11-hecc5488_0
certifi anaconda --> conda-forge
openssl anaconda::openssl-1.1.1-h7b6447c_0 --> conda-forge::openssl-1.1.1d-h516909a_0
r-ggplot2 conda-forge/noarch::r-ggplot2-3.2.1-r~ --> conda-forge/linux-64::r-ggplot2-2.2.1-r351h6115d3f_1
r-pillar conda-forge/noarch::r-pillar-1.4.2-r3~ --> pkgs/r/linux-64::r-pillar-1.3.0-r351h6115d3f_0
The following packages will be DOWNGRADED:
bioconductor-anno~ 1.62.0-r36_1 --> 1.60.1-r351_0
bioconductor-anno~ 1.46.0-r36_1 --> 1.44.0-r351_0
bioconductor-biob~ 2.44.0-r36h516909a_1 --> 2.42.0-r351h14c3975_1
bioconductor-bioc~ 0.30.0-r36_1 --> 0.28.0-r351_1
bioconductor-bioc~ 1.18.0-r36he1b5a44_1 --> 1.16.6-r351h1c2f66e_0
bioconductor-biom~ 2.40.3-r36_0 --> 2.38.0-r351_0
bioconductor-bios~ 2.52.0-r36h516909a_1 --> 2.50.2-r351h14c3975_0
bioconductor-biov~ 1.32.0-r36h516909a_1 --> 1.18.0-1
bioconductor-bsge~ 1.52.0-r36_1 --> 1.50.0-r351_0
bioconductor-dela~ 0.10.0-r36h516909a_1 --> 0.8.0-r351h14c3975_0
bioconductor-deseq 1.36.0-r36h516909a_1 --> 1.34.1-r351h14c3975_0
bioconductor-dese~ 1.24.0-r36he1b5a44_1 --> 1.22.1-r351hf484d3e_0
bioconductor-edger 3.26.5-r36he1b5a44_0 --> 3.26.0-r351hf484d3e_0
bioconductor-gene~ 1.66.0-r36hc99cbb1_1 --> 1.64.0-r351h1c2f66e_1
bioconductor-gene~ 1.62.0-r36_1 --> 1.60.0-r351_0
bioconductor-geno~ 1.20.0-r36_1 --> 1.18.1-r351_0
bioconductor-geno~ 1.2.1-r36_1 --> 1.2.1-r351_0
bioconductor-geno~ 1.20.1-r36h516909a_0 --> 1.18.1-r351h14c3975_0
bioconductor-geno~ 1.36.4-r36_0 --> 1.34.1-r351_0
bioconductor-geno~ 1.36.0-r36h516909a_1 --> 1.34.0-r351h14c3975_0
bioconductor-hpar 1.26.0-r36_1 --> 1.26.0-r351_0
bioconductor-iran~ 2.18.2-r36h516909a_0 --> 2.16.0-r351h14c3975_0
bioconductor-limma 3.40.2-r36h516909a_0 --> 3.40.0-r351h14c3975_0
bioconductor-rsam~ 2.0.0-r36he1b5a44_1 --> 1.34.0-r351hf484d3e_0
bioconductor-rtra~ 1.44.2-r36h516909a_1 --> 1.42.1-r351h9d9f1b6_1
bioconductor-s4ve~ 0.22.0-r36h516909a_1 --> 0.20.1-r351h14c3975_0
bioconductor-summ~ 1.14.0-r36_1 --> 1.12.0-r351_0
bioconductor-vari~ 1.30.1-r36h516909a_0 --> 1.28.3-r351h14c3975_0
bioconductor-xvec~ 0.24.0-r36h516909a_1 --> 0.22.0-r351h14c3975_0
bioconductor-zlib~ 1.30.0-r36h516909a_1 --> 1.28.0-r351h14c3975_0
r-acepack 1.4.1-r36h9bbef5b_1004 --> 1.4.1-r35h9bbef5b_1004
r-assertthat 0.2.1-r36h6115d3f_1 --> 0.2.1-r35h6115d3f_1
r-backports 1.1.5-r36hcdcec82_0 --> 1.1.5-r35hcdcec82_0
r-base 3.6.1-h8900bf8_2 --> 3.5.1-h08e1455_1008
r-base64enc 0.1_3-r36hcdcec82_1003 --> 0.1_3-r35hcdcec82_1003
r-bh 1.69.0_1-r36h6115d3f_1 --> 1.69.0_1-r35h6115d3f_1
r-bit 1.1_14-r36hcdcec82_1 --> 1.1_14-r35hcdcec82_1
r-bit64 0.9_7-r36hcdcec82_1001 --> 0.9_7-r35hcdcec82_1001
r-bitops 1.0_6-r36hcdcec82_1003 --> 1.0_6-r35hcdcec82_1003
r-blob 1.2.0-r36_1 --> 1.1.1-r351_1001
r-catools 1.17.1.2-r36h0357c0b_1 --> 1.17.1.2-r35h0357c0b_1
r-checkmate 1.9.4-r36hcdcec82_1 --> 1.9.4-r35hcdcec82_1
r-cli 1.1.0-r36h6115d3f_2 --> 1.1.0-r35h6115d3f_2
r-cluster 2.1.0-r36h9bbef5b_2 --> 2.1.0-r35h9bbef5b_2
r-colorspace 1.4_1-r36hcdcec82_1 --> 1.4_1-r35hcdcec82_1
r-crayon 1.3.4-r36h6115d3f_1002 --> 1.3.4-r35h6115d3f_1002
r-crosstalk 1.0.0-r36h6115d3f_1002 --> 1.0.0-r35h6115d3f_1002
r-curl 4.2-r36hcdcec82_0 --> 4.2-r35hcdcec82_0
r-data.table 1.12.6-r36hcdcec82_0 --> 1.12.6-r35hcdcec82_0
r-dbi 1.0.0-r36h6115d3f_1002 --> 1.0.0-r35h6115d3f_1002
r-dichromat 2.0_0-r36_2001 --> 2.0_0-r35_2001
r-digest 0.6.22-r36h0357c0b_0 --> 0.6.22-r35h0357c0b_0
r-dt 0.9-r36h6115d3f_0 --> 0.9-r35h6115d3f_0
r-evaluate 0.14-r36h6115d3f_1 --> 0.14-r35h6115d3f_1
r-fansi 0.4.0-r36hcdcec82_1001 --> 0.4.0-r35hcdcec82_1001
r-fastcluster 1.1.25-r36hc99cbb1_1002 --> 1.1.25-r35hc99cbb1_1002
r-foreign 0.8_72-r36hcdcec82_0 --> 0.8_72-r35hcdcec82_0
r-formatr 1.7-r36h6115d3f_1 --> 1.7-r35h6115d3f_1
r-formula 1.2_3-r36h6115d3f_1002 --> 1.2_3-r35h6115d3f_1002
r-futile.logger 1.4.3-r36h6115d3f_1002 --> 1.4.3-r35h6115d3f_1002
r-futile.options 1.0.1-r36h6115d3f_1001 --> 1.0.1-r35h6115d3f_1001
r-gdata 2.18.0-r36h6115d3f_1002 --> 2.18.0-r35h6115d3f_1002
r-glue 1.3.1-r36hcdcec82_1 --> 1.3.1-r35hcdcec82_1
r-gmd 0.3.3-r36h516909a_1002 --> 0.3.3-r35h516909a_1002
r-gplots 3.0.1.1-r36h6115d3f_1 --> 3.0.1.1-r35h6115d3f_1
r-gridextra 2.3-r36h6115d3f_1002 --> 2.3-r35h6115d3f_1002
r-gsalib 2.1-r36_1001 --> 2.1-r35_1001
r-gtable 0.3.0-r36h6115d3f_2 --> 0.3.0-r35h6115d3f_2
r-gtools 3.8.1-r36hcdcec82_1003 --> 3.8.1-r35hcdcec82_1003
r-highr 0.8-r36h6115d3f_1 --> 0.8-r35h6115d3f_1
r-hmisc 4.2_0-r36h9bbef5b_2 --> 4.2_0-r35h9bbef5b_2
r-hms 0.5.2-r36h6115d3f_0 --> 0.4.2-r351h6115d3f_1000
r-htmltable 1.13.2-r36h6115d3f_0 --> 1.13.2-r35h6115d3f_0
r-htmltools 0.4.0-r36h0357c0b_0 --> 0.4.0-r35h0357c0b_0
r-htmlwidgets 1.5.1-r36h6115d3f_0 --> 1.5.1-r35h6115d3f_0
r-httpuv 1.5.2-r36h0357c0b_1 --> 1.5.2-r35h0357c0b_1
r-httr 1.4.1-r36h6115d3f_1 --> 1.4.1-r35h6115d3f_1
r-jsonlite 1.6-r36hcdcec82_1001 --> 1.6-r35hcdcec82_1001
r-kernsmooth 2.23_15-r36h9bbef5b_1004 --> 2.23_15-r35h9bbef5b_1004
r-knitr 1.25-r36h6115d3f_0 --> 1.25-r35h6115d3f_0
r-labeling 0.3-r36h6115d3f_1002 --> 0.3-r35h6115d3f_1002
r-lambda.r 1.2.4-r36h6115d3f_0 --> 1.2.4-r35h6115d3f_0
r-later 1.0.0-r36h0357c0b_0 --> 1.0.0-r35h0357c0b_0
r-lattice 0.20_38-r36hcdcec82_1002 --> 0.20_38-r35hcdcec82_1002
r-latticeextra 0.6_28-r36h6115d3f_1002 --> 0.6_28-r35h6115d3f_1002
r-lazyeval 0.2.2-r36hcdcec82_1 --> 0.2.2-r35hcdcec82_1
r-locfit 1.5_9.1-r36h516909a_1004 --> 1.5_9.1-r35h516909a_1004
r-magrittr 1.5-r36h6115d3f_1002 --> 1.5-r35h6115d3f_1002
r-markdown 1.1-r36hcdcec82_0 --> 1.1-r35hcdcec82_0
r-mass 7.3_51.4-r36hcdcec82_1 --> 7.3_51.4-r35hcdcec82_1
r-matrix 1.2_17-r36hcdcec82_1 --> 1.2_17-r35hcdcec82_1
r-matrixstats 0.55.0-r36hcdcec82_0 --> 0.55.0-r35hcdcec82_0
r-memoise 1.1.0-r36h6115d3f_1003 --> 1.1.0-r35h6115d3f_1003
r-mime 0.7-r36hcdcec82_1 --> 0.7-r35hcdcec82_1
r-munsell 0.5.0-r36h6115d3f_1002 --> 0.5.0-r35h6115d3f_1002
r-nnet 7.3_12-r36hcdcec82_1003 --> 7.3_12-r35hcdcec82_1003
r-openssl 1.4.1-r36h9c8475f_0 --> 1.1-r351h3a9d887_1002
r-pkgconfig 2.0.3-r36h6115d3f_0 --> 2.0.3-r35h6115d3f_0
r-plogr 0.2.0-r36h6115d3f_1002 --> 0.2.0-r35h6115d3f_1002
r-plotrix 3.7_6-r36h6115d3f_1 --> 3.7_6-r35h6115d3f_1
r-plyr 1.8.4-r36h0357c0b_1003 --> 1.8.4-r35h0357c0b_1003
r-prettyunits 1.0.2-r36h6115d3f_1002 --> 1.0.2-r35h6115d3f_1002
r-progress 1.2.2-r36h6115d3f_1 --> 1.2.2-r35h6115d3f_1
r-promises 1.1.0-r36h0357c0b_0 --> 1.1.0-r35h0357c0b_0
r-r6 2.4.0-r36h6115d3f_2 --> 2.4.0-r35h6115d3f_2
r-rcolorbrewer 1.1_2-r36h6115d3f_1002 --> 1.1_2-r35h6115d3f_1002
r-rcpparmadillo 0.9.800.1.0-r36h0357c0b_0 --> 0.9.800.1.0-r35h0357c0b_0
r-rcurl 1.95_4.12-r36hcdcec82_1 --> 1.95_4.12-r35hcdcec82_1
r-reshape 0.8.8-r36hcdcec82_1 --> 0.8.8-r35hcdcec82_1
r-reshape2 1.4.3-r36h0357c0b_1004 --> 1.4.3-r35h0357c0b_1004
r-rlang 0.4.1-r36hcdcec82_0 --> 0.4.1-r35hcdcec82_0
r-rmarkdown 1.16-r36h6115d3f_0 --> 1.16-r35h6115d3f_0
r-rpart 4.1_15-r36hcdcec82_1 --> 4.1_15-r35hcdcec82_1
r-rsqlite 2.1.2-r36h0357c0b_1 --> 2.1.2-r35h0357c0b_1
r-rstudioapi 0.10-r36h6115d3f_2 --> 0.10-r35h6115d3f_2
r-scales 1.0.0-r36h0357c0b_1002 --> 1.0.0-r35h0357c0b_1002
r-shiny 1.4.0-r36h6115d3f_0 --> 1.3.2-r35h6115d3f_1
r-snow 0.4_3-r36h6115d3f_1001 --> 0.4_3-r35h6115d3f_1001
r-sourcetools 0.1.7-r36he1b5a44_1001 --> 0.1.7-r35he1b5a44_1001
r-stringi 1.4.3-r36h0357c0b_2 --> 1.4.3-r35h0357c0b_2
r-stringr 1.4.0-r36h6115d3f_1 --> 1.4.0-r35h6115d3f_1
r-survival 2.44_1.1-r36hcdcec82_1 --> 2.44_1.1-r35hcdcec82_1
r-tibble 2.1.3-r36hcdcec82_1 --> 1.4.2-r351h96ca727_1002
r-utf8 1.1.4-r36hcdcec82_1001 --> 1.1.4-r35hcdcec82_1001
r-venndiagram 1.6.20-r36h6115d3f_1001 --> 1.6.20-r35h6115d3f_1001
r-viridis 0.5.1-r36h6115d3f_1003 --> 0.5.1-r35h6115d3f_1003
r-viridislite 0.3.0-r36h6115d3f_1002 --> 0.3.0-r35h6115d3f_1002
r-xfun 0.10-r36h6115d3f_0 --> 0.10-r35h6115d3f_0
r-xml 3.98_1.20-r36hcdcec82_1 --> 3.98_1.20-r35hcdcec82_1
r-xtable 1.8_4-r36h6115d3f_2 --> 1.8_4-r35h6115d3f_2
r-yaml 2.2.0-r36hcdcec82_1002 --> 2.2.0-r35hcdcec82_1002
In the meantime, I suggest you have update your
~/.condarc
with:- conda-forge - bioconda - defaults
Also, if you have a recent re-install, could you please share
conda list rpy2
Updated - by update I guess you mean to remove the existing depositories, which I have done.
Sure, that's what I meant.
Ok, let me do some testing and I will come back to you.
Re: rpy2, I can see that you have version 3.1.0, which was installed with the installer but according to Jenkins that doesn't pass the tests. I can share details with you via PM.
Aware of at least some of the rpy2 issues, recently pushed a compatibility fix to make both versions of rpy2 work. They appare to have moved the RRuntime library to another place.
Hmm just a slightly more philosophical suggestion from me to tackle some of these recurrent problems - feel free to ignore.
That said, the R tximport package would be a must to be supported, as currently the salmon -> tximport -> deseq2 is stil a widely used and preferred rnaseqdiffexpression workflow?
I agree with you, many thanks for sharing your thoughts!
Very keen on improving things as much as possible.
Re: r-wasaby. We do have separate conda environments for special tools that won't be updated but are required for some pipelines to work. We could do the same for r-wasabi if you think that's the best approach.
Hi Sebastian, ran another fresh install in the background. Failed at the salmon stage.
The code successfully installed!
To activate the CGAT environment type:
$ source /ifs/projects/jakubs/cgat-developers-v0/conda-install/etc/profile.d/conda.sh
$ conda activate base
$ conda activate cgat-flow
To deactivate the environment, use:
$ conda deactivate
# install-devel.sh log | cgath1.anat.ox.ac.uk | Tue Nov 12 17:15:15 GMT 2019 | install pipeline deps
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1608 100 1608 0 0 8288 0 --:--:-- --:--:-- --:--:-- 8246
Collecting package metadata (repodata.json): done
Solving environment: |
Found conflicts! Looking for incompatible packages.
This can take several minutes. Press CTRL-C to abort.
failed
UnsatisfiableError: The following specifications were found to be incompatible with each other:
Package llvm-meta conflicts for:
salmon -> libcxx -> llvm-meta[version='6.0.1|7.0.0']
Exactly, that's the current issue.
When we leave rpy2 unpinned, version 3.1.0 is installed and tests fail with:
ImportError: cannot import name 'process_revents'
I haven't looked into why is that.
If we pin rpy2 to <= 2.9, then the installation fails with the error you just pasted. I think it's trying to install the latest salmon (e.g. version 1.0.0) and there is a conflict.
Either we solve the rpy2 version 3.1.0 issue, or we work with a previous version of salmon (e.g. <= 0.14)
Any preferences?
Moving rpy2 to version 3 would be a significant rewrite of the codebase. Verison 3 included a complete rewrite of the interface. https://rpy2.github.io/doc/v3.0.x/html/changes.html
Example error: Pandas - we have lots of these https://stackoverflow.com/questions/55990529/module-rpy2-robjects-pandas2ri-has-no-attribute-ri2py https://rpy2.github.io/doc/v3.0.x/html/generated_rst/pandas.html
The only problem with installing earlier version of salmon would be that I have experienced segmentation faults when running on the cluster. However, I couldn't work out the exact reason for these seg faults.
However, there is no reason to freeze the entire conda environment in the stone age because of one package. That would have a negative impact on research. The two options are:
With respect to this branch I have fixed what I could. The pipeline now runs almost to the end again. Apart from the refactoring (completely changed geneset creation and removed sailfish) I have also detected a significant number of minor errors which resulted in errors in some of the minor tasks.
The remaining issues are:
runSalmonSaturation
fails because salmon recognises the truncated highest depth samples as corrupt.loadSalmonResults
now loads a "wide" table with columns representing a variable. The previous table was a "narrow table", and this has an impact on plotTopGenesHeatmap
and plotExpression
which were written for the narrow table (like the tables used by ggplot2).Further testing is limited by my version of rpy2 (3.1.0).
Hi,
I have been playing a little bit more with conda and https://github.com/cgat-developers/cgat-flow/commit/82791ccc780083821b0812f0dde6df6822d55207 seems to make the trick.
It now installs salmon version 1.0.0 and rpy2 2.9.4, which should be what we need.
@jscaber if you re-install, you should get the correct environment now. Instead of re-installing, you could also try conda install 'rpy2 < 3'
and check whether that works.
Best regards, Sebastian
Thanks Sebastian.
This is an example of what changes for rpy2 v3 compatibility might look like, not too terrible.
from rpy2.robjects import pandas2ri
+import rpy2.robjects as ro
+from rpy2.robjects.conversion import localconverter
import cgatcore.experiment as E
import cgat.GTF as GTF
@@ -1420,15 +1425,17 @@ def plotTopGenesHeatmap(outfile):
- plotHeatmap(pandas2ri.py2ri(intersection_pivot),
- pandas2ri.py2ri(factors_df))
+ with localconverter(ro.default_converter + pandas2ri.converter):
+ r_intersection_pivot = ro.conversion.py2rpy(intersection_pivot)
+ r_factors_df = ro.conversion.py2rpy(factors_df)
+
+ plotHeatmap(r_intersection_pivot,
+ r_factors_df)
iotools.touch_file(outfile)
Thanks, Jakub.
Indeed, it doesn't look terrible. However, I don't know how to quantify the effort for the remaining refactoring to migrate the code to rpy2 v3, and more importantly, I won't be able to help you with that particular task. So it is up to you guys (the end user of the production pipelines) to decide what's the best decision going forward.
Best regards, Sebastian
Following up on the r-wasabi
install, it depends on R version 3.5, and our main cgat-flow
environment installs R version 3.6. Hence the large amount of changes required to install it.
If we want to keep r-wasabi
as a dependency, I suggest we use it in a separate conda environment like other outdated dependencies.
Please let me know your thoughts.
Hi,
Thank you @sebastian-luna-valero for all your hard work on this. I will speak to Adam about wasabi - but removing or creating a separate conda environment is the way to go.
Regarding rpy2, it looks like 2.9.4 is a little intermediate.
While the pandas2ri.py2ri
did not fail, the implicit conversions it induces were not tolerated. I think I could have overcome this with .activate()
, but chose instead to use the explicit local converters: they have the same syntax as rpy2 v3 with one exception: ro.conversion.py2rpy
in 3.1 is ro.conversion.py2ri
2.9.4! The mainers of rpy2 clearly love refactoring code.
I have fixed all the plot functions here and the only thing that does not work is runSalmonSaturation
: this fails because the truncated bam of different sizes is recognised by salmon as corrupt. This may be something with my .bam though and I am not inclined to investgate.
@Acribbs could you test this branch? (I promise I will set up my own testing environment soon, too)
Also, for the plotExpressionDensity I had to change the log2(TPM + 0.1) to log2(TPM + 1)
to avoid negative values. Is there a reason it was 0.1 - I may be missing something?
Hi,
I have been playing a little bit more with conda and 82791cc seems to make the trick.
It now installs salmon version 1.0.0 and rpy2 2.9.4, which should be what we need.
@jscaber if you re-install, you should get the correct environment now. Instead of re-installing, you could also try
conda install 'rpy2 < 3'
and check whether that works.Best regards, Sebastian
Btw conda install 'rpy2 < 3'
worked. No conflicts.
Hi @jscaber im going to be travelling for a few weeks but should be able to test this in-between trips. I will let you know outcome
Hi,
Could we please recap what's needed to close this issue?
Best regards, Sebastian
I forgot about this, I will test this branch sing pipeline testing framework and get back to you.
Have tested this and all seems to be fine.
ping @jscaber
@jscaber the testing pipeline needs updating and I was thinking of making a separate branch to do this. However, I started on this branch. I will merge this and make a separate branch