pipemg / kcone-paper Goto Github PK

dycone has now seen various updates since the original paper manuscript. The repository here provides the protocol to reproduce the paper with the most recent version of dycone.

If you want to reproduce the paper with the same version of dycone as used originally please see the legacy protocol to be found at https://cdiener.github.io/kcone-paper/legacy.html. For that you will need to install the legacy version of dycone with:

devtools::install_github("cdiener/dycone", ref = "legacy")

The manuscript is accepted as: "The space of enzyme regulation in HeLa cells can be inferred from its intracellular metabolome" Christian Diener, Felipe Muñoz-Gonzalez, Sergio Encarnación, Osbaldo Resendis-Antonio

This repository contains all materials in order to reproduce all results from our publication. You can either create a detailed computational protocol from an Rmarkdown file or just run an R script to reproduce the results.

The output can be found at https://cdiener.github.io/kcone-paper.

For the source code of the dycone package please see https://github.com/cdiener/dycone.

Please use the following paper: http://www.ncbi.nlm.nih.gov/pubmed/27335086.

This repository provides the following input files:

• metabolome measurements for HaCaT and HeLa (metabolome.csv)
• model for central carbon metabolism with annotations (reactions.csv)
• a mapping of the model metabolites to KEGG compound and HMDB IDs (id_map.csv)
• a list of 58 microrray samples from the GEO database annotated with the corresponding cell line (ge_samples.csv)
• the model in dycone's CSV format (reactions.csv)
• the model in SBML (L3V1) format (cemet.xml)

Additionally, upon first run we will generate some intermediate data files that include results for the most time consuming steps in the analysis. This way the first run will take about 3h on a machine with 6+ cores, but repeated runs will finish in about 4 minutes. The intermediate files cover the following protocol steps:

• web scraping of missing metabolite concentrations from HMDB (scraped_concs.Rd)
• calculation of the k-cone basis for the model (basis.Rd)
• stability analysis for all of the 6 bases (stab.Rd)

For the sake of completeness we also provide the pathway image we use in the paper as Inkscape SVGs in the images folder. This image is licensed under the CC-BY 4.0.

In order create HTML and PDF outputs of the protocol follow the initial installation intructions (section "Installation")and render the entire Rmarkdown protocol with

Rscript -e "rmarkdown::render('protocol.Rmd', output_format='all')"

HTML output is the same as found on https://cdiener.github.io/kcone-paper.

In case you want to run the analysis in a "quiet" manner without producing the HTML and PDF reports but yielding all other output (tables and figure files) you can also run the script version via

Rscript -e "source('protocol.R')"

This is useful if you want to perform small changes and see how this alters the results.