Very different results with latest CARNIVAL release

[laylagerami]

Hi devs,

I have noticed that I am getting very different results with CARNIVAL now compared to the previous version. I am using the exact same OS, cplex version, input data and settings (default w/ 20 cores).

Specifically, before the network I obtained had 69 nodes and 140 edges. Running with the new version of CARNIVAL, the weighted network now has 2535 nodes and 7672 edges. This is quite large since I would like to visualise the network in a Shiny app! Is this behaviour intentional?

I also tried running a different dataset with both cbc+cplex and again obtained very big networks compared to what I was getting before, but haven't got an exact "before" to compare to.

Command before update

values$carnival_result <- runCARNIVAL(
          netObj = pkn,
          measObj = meas,
          weightObj = weights,
          solverPath = "/scratch/lh605/ucc-fs-nethome/ibm/ILOG/CPLEX_Studio1210/cplex/bin/x8664_linux/cplex",
          solver = "cplex",
          timelimit = 3600,
          threads = 20
       )

Command upon update

carnival_options = defaultCplexCarnivalOptions()
      carnival_options$solverPath = "/scratch/lh605/ucc-fs-nethome/ibm/ILOG/CPLEX_Studio1210/cplex/bin/x8664_linux/cplex"
      carnival_options$timelimit = 3600
      carnival_options$threads = 20
      carnival_options$workdir = "output/"
      carnival_options$outputFolder = "output/"

values$carnival_result <- runInverseCarnival(
          priorKnowledgeNetwork = pkn,
          measurements = meas,
          weights = weights,
          carnivalOptions = carnival_options
          )

Session info:

R version 4.1.0 (2021-05-18)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 18.04.5 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/openblas/libblas.so.3
LAPACK: /usr/lib/x86_64-linux-gnu/libopenblasp-r0.2.20.so

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

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

other attached packages:
 [1] readr_1.4.0          snowfall_1.84-6.1    snow_0.4-3           GSEABase_1.54.0     
 [5] graph_1.70.0         annotate_1.70.0      XML_3.99-0.6         colorspace_2.0-1    
 [9] sortable_0.4.4       lpSolve_5.6.15       shinyWidgets_0.6.0   shinyalert_2.0.0    
[13] HGNChelper_0.8.1     piano_2.8.0          visNetwork_2.0.9     CARNIVAL_2.2.0      
[17] progeny_1.14.0       ggplot2_3.3.3        tibble_3.1.2         dplyr_1.0.6         
[21] dorothea_1.4.0       org.Hs.eg.db_3.13.0  AnnotationDbi_1.54.0 IRanges_2.26.0      
[25] S4Vectors_0.30.0     Biobase_2.52.0       BiocGenerics_0.38.0  shinyFiles_0.9.0    
[29] shinythemes_1.2.0    shinyBS_0.61         shinysky_0.1.3       rhandsontable_0.3.8 
[33] chemdoodle_0.4.0     rcdk_3.5.0           rcdklibs_2.3         rJava_1.0-4         
[37] miniUI_0.1.1.1       DT_0.18              igraph_1.2.6         shinyjs_2.0.0       
[41] shiny_1.6.0         

loaded via a namespace (and not attached):
  [1] uuid_0.1-4             fastmatch_1.1-0        plyr_1.8.6             fingerprint_3.5.7     
  [5] shinydashboard_0.7.1   splines_4.1.0          BiocParallel_1.26.0    crosstalk_1.1.1       
  [9] GenomeInfoDb_1.28.0    digest_0.6.27          htmltools_0.5.1.1      fansi_0.5.0           
 [13] magrittr_2.0.1         memoise_2.0.0          cluster_2.1.2          mixtools_1.2.0        
 [17] limma_3.48.0           Biostrings_2.60.0      blob_1.2.1             ggrepel_0.9.1         
 [21] xfun_0.23              crayon_1.4.1           RCurl_1.98-1.3         jsonlite_1.7.2        
 [25] survival_3.2-11        iterators_1.0.13       glue_1.4.2             gtable_0.3.0          
 [29] zlibbioc_1.38.0        XVector_0.32.0         kernlab_0.9-29         scales_1.1.1          
 [33] DBI_1.1.1              relations_0.6-9        Rcpp_1.0.6             xtable_1.8-4          
 [37] learnr_0.10.1          bit_4.0.4              proxy_0.4-25           htmlwidgets_1.5.3     
 [41] httr_1.4.2             fgsea_1.18.0           gplots_3.1.1           ellipsis_0.3.2        
 [45] farver_2.1.0           pkgconfig_2.0.3        sass_0.4.0             utf8_1.2.1            
 [49] RJSONIO_1.3-1.4        labeling_0.4.2         tidyselect_1.1.1       rlang_0.4.11          
 [53] later_1.2.0            munsell_0.5.0          tools_4.1.0            cachem_1.0.5          
 [57] cli_2.5.0              generics_0.1.0         RSQLite_2.2.7          stringr_1.4.0         
 [61] evaluate_0.14          fastmap_1.1.0          yaml_2.2.1             knitr_1.33            
 [65] bit64_4.0.5            fs_1.5.0               bcellViper_1.28.0      caTools_1.18.2        
 [69] purrr_0.3.4            KEGGREST_1.32.0        mime_0.10              slam_0.1-48           
 [73] compiler_4.1.0         rstudioapi_0.13        png_0.1-7              e1071_1.7-7           
 [77] marray_1.70.0          viper_1.26.0           stringi_1.6.2          bslib_0.2.5.1         
 [81] lattice_0.20-44        Matrix_1.3-4           markdown_1.1           vctrs_0.3.8           
 [85] pillar_1.6.1           lifecycle_1.0.0        jquerylib_0.1.4        data.table_1.14.0     
 [89] bitops_1.0-7           httpuv_1.6.1           R6_2.5.0               promises_1.2.0.1      
 [93] KernSmooth_2.23-20     gridExtra_2.3          MASS_7.3-54            gtools_3.8.2          
 [97] assertthat_0.2.1       rprojroot_2.0.2        withr_2.4.2            GenomeInfoDbData_1.2.6
[101] hms_1.1.0              grid_4.1.0             tidyr_1.1.3            class_7.3-19          
[105] rmarkdown_2.8          segmented_1.3-4        sets_1.0-18            itertools_0.1-3  

Let me know if you need anything else from my side to understand what's going on,

Layla

[ivanovaos]

Hi Layla,

Sorry for the delayed response. There was indeed an exporting solution bug in the new version. Can you check if the results are similar now?

[laylagerami]

Hi! I was about to check it but I had an issue re-installing the package from GitHub.

Malformed Depends or Suggests or Imports or Enhances field.
   Offending entries:
     plyr
   rjson
   Entries must be names of packages optionally followed by '<=' or '>=',
   white space, and a valid version number in parentheses.

I think this is due to a comma missing after "plyr" in DESCRIPTION file, see below:

Imports:
    readr,
    stringr,
    lpSolve,
    igraph,
    dplyr,
    plyr <- (here)
    rjson,
    rmarkdown

[ivanovaos]

Can you try once more?

[laylagerami]

Thanks - running now. Will let you know in an hour or so!

EDIT: Sorry, my Rstudio server is crashing (OOO and connection is unstable here also w/ VPN). I will take a look on Monday when I'm back home and let you know.

[laylagerami]

Network size as expected now -- thanks!