Development Stage: Beta
CoRC, the COPASI R Connector, links the Complex Pathway Simulator COPASI (copasi.org) and the (statistical) programming environment R (r-project.org). It provides easy access to the powerful biochemical model editing, simulation and analysis backend of COPASI from the convenient R command line interface. This allows the user to develop elaborate scripts and workflows for analyses that would require a great deal of tedious manual work otherwise. These scripts can then be run interactively from the R command line interface or be sent to cluster or cloud facilities for more demanding calculations.
CoRC features:
- high-level API for COPASI in the R language.
- Immediate access to R’s data analysis capabilities and publication-ready plotting.
- Reproducible workflows from data generation to analysis and plotting (R scripts and notebooks).
- Rule-based modification of model structure to test structural variations or create large models.
- Scaling up assays, e.g. from 3 models to 3000.
- Handling of multiple models at once.
- Parallelization on multi-core machines or computing clusters.
It is currently based on COPASI version 4.39 Build (272) and aims to closely follow COPASI releases in the future.
CoRC comes with the Artistic License 2.0. By using CoRC you agree to this license.
Install the CoRC package directly from GitHub:
install.packages("remotes")
remotes::install_github("jpahle/CoRC")
The installation may take a few minutes.
CoRC runs it’s own version of the COPASI backend which it will download for your platform during installation.
library(CoRC)
loadExamples(1)
#> [[1]]
#> # A COPASI model reference:
#> Model name: "The Brusselator"
#> Number of compartments: 1
#> Number of species: 6
#> Number of reactions: 4
runTimeCourse()$result
#> # A tibble: 201 × 3
#> Time X Y
#> <dbl> <dbl> <dbl>
#> 1 0 3 3
#> 2 0.5 3.41 0.817
#> 3 1 1.90 1.28
#> 4 1.5 0.876 1.87
#> 5 2 0.346 2.37
#> 6 2.5 0.183 2.66
#> 7 3 0.147 2.86
#> 8 3.5 0.141 3.05
#> 9 4 0.141 3.23
#> 10 4.5 0.142 3.41
#> # … with 191 more rows