pavlidislab / erminer Goto Github PK

these work a little differently and redundant

Hey Ogan,

I am still having issues running ermineR functions. To try and solve the problem with the annotation file, I thought I could create my own file and then use it. But I kept getting the below error:

Error in (function (annotation = NULL, aspects = c("Molecular Function", : Something went wrong. Blame the dev sh: /Library/Frameworks/R.framework/Versions/4.1/Resources/library/ermineR/ermineJ-3.2/bin/ermineJ.sh: Permission denied

Any help would be greatly appreciated!
Thanks

They have multiple java installations. It finds the wrong one

Hello,

In the past I have installed ermineR using install_github() successfully. Today I tried installing it on a different machine and it produced an error:

devtools::install_github('PavlidisLab/ermineR')
Downloading GitHub repo PavlidisLab/ermineR@master
Erreur : Failed to install 'ermineR' from GitHub:
HTTP error 404.
Not Found

Did you spell the repo owner (oganm) and repo name (gemmaAPI.R) correctly?

• If spelling is correct, check that you have the required permissions to access the repo.

I noticed there was a commit 29 days ago regarding gemmaAPI and was wondering whether this may be related to a permissions issue?

Thanks!

Session info:

R version 3.6.2 (2019-12-12)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: CentOS Linux 7 (Core)

Matrix products: default
BLAS: /usr/local/R-3.6.2/lib64/R/lib/libRblas.so
LAPACK: /usr/local/R-3.6.2/lib64/R/lib/libRlapack.so

locale:
[1] LC_CTYPE=fr_FR.UTF-8 LC_NUMERIC=C LC_TIME=fr_FR.UTF-8
[4] LC_COLLATE=fr_FR.UTF-8 LC_MONETARY=fr_FR.UTF-8 LC_MESSAGES=fr_FR.UTF-8
[7] LC_PAPER=fr_FR.UTF-8 LC_NAME=C LC_ADDRESS=C
[10] LC_TELEPHONE=C LC_MEASUREMENT=fr_FR.UTF-8 LC_IDENTIFICATION=C

attached base packages:
[1] stats4 parallel stats graphics grDevices utils datasets methods base

other attached packages:
[1] gridExtra_2.3 comeback_0.1.0
[3] robCompositions_2.2.1 data.table_1.12.8
[5] pls_2.7-2 methylumi_2.32.0
[7] FDb.InfiniumMethylation.hg19_2.2.0 org.Hs.eg.db_3.10.0
[9] TxDb.Hsapiens.UCSC.hg19.knownGene_3.2.2 GenomicFeatures_1.38.2
[11] AnnotationDbi_1.48.0 scales_1.1.0
[13] SDMTools_1.1-221.2 devtools_2.2.1
[15] usethis_1.5.1 heatmap.plus_1.3
[17] RColorBrewer_1.1-2 DescTools_0.99.32
[19] reshape2_1.4.3 IlluminaHumanMethylationEPICmanifest_0.3.0
[21] ggplot2_3.2.1 IHW_1.14.0
[23] dplyr_0.8.4 minfi_1.32.0
[25] bumphunter_1.28.0 locfit_1.5-9.1
[27] iterators_1.0.12 foreach_1.4.8
[29] Biostrings_2.54.0 XVector_0.26.0
[31] SummarizedExperiment_1.16.1 DelayedArray_0.12.2
[33] BiocParallel_1.20.1 matrixStats_0.55.0
[35] Biobase_2.46.0 GenomicRanges_1.38.0
[37] GenomeInfoDb_1.22.0 IRanges_2.20.2
[39] S4Vectors_0.24.3 BiocGenerics_0.32.0

loaded via a namespace (and not attached):
[1] rappdirs_0.3.1 rtracklayer_1.46.0 prabclus_2.3-2 GGally_1.4.0
[5] R.methodsS3_1.8.0 tidyr_1.0.2 bit64_0.9-7 knitr_1.28
[9] R.utils_2.9.2 RCurl_1.98-1.1 GEOquery_2.54.1 preprocessCore_1.48.0
[13] callr_3.4.2 RSQLite_2.2.0 bit_1.1-15.2 xml2_1.2.2
[17] assertthat_0.2.1 xfun_0.12 hms_0.5.3 DEoptimR_1.0-8
[21] fansi_0.4.1 scrime_1.3.5 progress_1.2.2 dbplyr_1.4.2
[25] readxl_1.3.1 DBI_1.1.0 reshape_0.8.8 purrr_0.3.3
[29] ellipsis_0.3.0 backports_1.1.5 annotate_1.64.0 sROC_0.1-2
[33] biomaRt_2.42.0 vctrs_0.2.3 remotes_2.1.1 abind_1.4-5
[37] withr_2.1.2 robustbase_0.93-5 vcd_1.4-5 GenomicAlignments_1.22.1
[41] fdrtool_1.2.15 prettyunits_1.1.1 mclust_5.4.5 cluster_2.1.0
[45] lazyeval_0.2.2 laeken_0.5.1 crayon_1.3.4 genefilter_1.68.0
[49] pkgconfig_2.0.3 zCompositions_1.3.3-1 slam_0.1-47 labeling_0.3
[53] nlme_3.1-144 pkgload_1.0.2 nnet_7.3-12 rlang_0.4.4
[57] diptest_0.75-7 lifecycle_0.1.0 BiocFileCache_1.10.2 cellranger_1.1.0
[61] rprojroot_1.3-2 lmtest_0.9-37 rngtools_1.5 base64_2.0
[65] Matrix_1.2-18 carData_3.0-3 lpsymphony_1.14.0 Rhdf5lib_1.8.0
[69] boot_1.3-24 zoo_1.8-7 processx_3.4.2 bitops_1.0-6
[73] R.oo_1.23.0 blob_1.2.1 DelayedMatrixStats_1.8.0 doRNG_1.8.2
[77] stringr_1.4.0 nor1mix_1.3-0 readr_1.3.1 memoise_1.1.0
[81] magrittr_1.5 plyr_1.8.5 zlibbioc_1.32.0 compiler_3.6.2
[85] illuminaio_0.28.0 rrcov_1.5-2 Rsamtools_2.2.2 cli_2.0.1
[89] ps_1.3.2 MASS_7.3-51.5 tidyselect_1.0.0 stringi_1.4.6
[93] forcats_0.4.0 yaml_2.2.1 askpass_1.1 grid_3.6.2
[97] tools_3.6.2 rio_0.5.16 rstudioapi_0.11 foreign_0.8-75
[101] farver_2.0.3 digest_0.6.25 quadprog_1.5-8 fpc_2.2-5
[105] Rcpp_1.0.3 car_3.0-6 siggenes_1.60.0 httr_1.4.1
[109] kernlab_0.9-29 colorspace_1.4-1 XML_3.99-0.3 fs_1.3.1
[113] ranger_0.12.1 truncnorm_1.0-8 splines_3.6.2 expm_0.999-4
[117] sp_1.3-2 multtest_2.42.0 flexmix_2.3-15 sessioninfo_1.1.1
[121] xtable_1.8-4 modeltools_0.2-22 testthat_2.3.1 R6_2.4.1
[125] NADA_1.6-1 pillar_1.4.3 htmltools_0.4.0 glue_1.3.1
[129] VIM_5.1.0 cvTools_0.3.2 class_7.3-15 beanplot_1.2
[133] codetools_0.2-16 pkgbuild_1.0.6 pcaPP_1.9-73 mvtnorm_1.0-12
[137] utf8_1.1.4 lattice_0.20-38 tibble_2.1.3 curl_4.3
[141] zip_2.0.4 openxlsx_4.1.4 openssl_1.4.1 survival_3.1-8
[145] limma_3.42.2 desc_1.2.0 munsell_0.5.0 e1071_1.7-3
[149] rhdf5_2.30.1 GenomeInfoDbData_1.2.2 HDF5Array_1.14.2 haven_2.2.0
[153] gtable_0.3.0

• Do annotation files have to be no-parents? YES
• What is -f "Directory where custom gene set are located"
• How to specify the Agilent platform type?

Reports include not having permission on install or permissions changing after an unspecified period of proper function.

To fix these issues manually adding the execution permissions to the jar in the installation file is the current solution

I have rJava installed and runned the chmod +x command but still wasn't able to get the pathway information that I wanted.

> oraOut <- ora(annotation = 'Generic_human', hitlist = hitlist)
Attempting to download annotation file
Error in (function (annotation = NULL, aspects = c("Molecular Function",  : 
  Something went wrong. Blame the dev
INFO: Data directory is /Users/garyzhubc/ermineJ.data
DEBUG: Custom gene sets directory is /Users/garyzhubc/ermineJ.data/genesets
DEBUG: Setting score start column to 2
INFO: Gene symbols for each term will be output
DEBUG: gene score threshold set to 0.5
Reading GO descriptions from /var/folders/w8/9_5f04c54b585y1kjq898d180000gn/T/RtmpCygyvd/file134597a7acf02 ... 
Reading gene annotations from /var/folders/w8/9_5f04c54b585y1kjq898d180000gn/T/RtmpCygyvd/file1345963d035c7 ... 
WARN: Duplicated element: NPHP3-ACAD11 at line 1435, skipping
Read 2500 elements ... 
WARN: Duplicated element: RNASEK-C17ORF49 at line 4697, skipping
Read 5000 elements ... 
WARN: Duplicated element: INO80B-WBP1 at line 5032, skipping
WARN: Duplicated element: C1QTNF3-AMACR at line 5148, skipping
WARN: Duplicated element: FLJ45825 at line 5673, skipping
Read 7500 elements ... 
WARN: Duplicated element: ARHGAP19-SLIT1 at line 7721, skipping
WARN: Duplicated element: DUX

ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/gene2go.gz

Good news is it was not hard to get working. I had to set JAVA_HOME with Sys.setenv, otherwise no show-stoppers - just regular bugs.

Some of these discussed previously.

1. Downloads GO file every time you run it?
2. Using a data frame for scores didn't work for me. The "scoreColumn" setting is also ambiguous between the two situations. Ideally we would only support data frames, and have a better way to index the scores (e.g., by column name, or just require feeding in a single column as a df with row names or list)
3. If possible make logging output less verbose
4. Top of help says logTrans=FALSE but help body says the default is TRUE. Is in fact FALSE by default.
5. Some of the options are pretty obscure and would not be used in normal use. For example the correlation option can be hidden (or would be a separate method)
6. Having simplified options like "ORA", "PreRe", "ROC" would be good rather than requiring test="GSR" and stats="precisionRecall".
7. Allow key diagnostics to be captured in the output object (as attributes I guess) - for example, the MF bias (correlation).

If any of these need changes on the ermineJ side, let me know. Also not all of these are important (depends on effort required) just nice to have.

I was trying to run

oraOut = ora(annotation = 'Generic_human',
             hitlist = hitlist)

And I am getting an error of:

cannot open URL 'http://release.geneontology.org/latest-termdb/go_daily-termdb.rdf-xml.gz': HTTP status was '404 Not Found'Error in utils::download.file("http://archive.geneontology.org/latest-termdb/go_daily-termdb.rdf-xml.gz",  : 
  cannot open URL 'http://archive.geneontology.org/latest-termdb/go_daily-termdb.rdf-xml.gz'

Is it possible to download the gene to function table without using the statistical procedure in the package?

For instance, it will be a table like:

       Function1 Function2

Gene1 x

Gene2 x x

that I would want.

findJava doesn't always work but it is not consistent.

CRAN can have packages with jar files in them. ErmineJ is over the recommended size of 5 mbs but that is not a hard limit. Tests that take too much time should be skip_on_cranned.

• Send R seed

• Fix names of multiple test correction methods

• Change default class limits

• Which terms to use is an input now