RHermes

RHermes is a broad-scoped targeted metabolomics software used to analyse LC-MS and LC-MS/MS data to identify compounds in biological or environmental samples.

The RHermes workflow works with both Orbitrap and q-TOF instrument data and comes with an easy to use GUI that will guide you every step of the way.

Check out the documentation here

System requirements

The recommended system requirements are:

  • At least a 4 core processor
  • 8-16 GB of RAM or more
  • An internet connection to perform KEGG queries

Installation

You can download the development version from GitHub with:

if(!requireNamespace("devtools", quietly = TRUE))
    install.packages("devtools")
devtools::install_github("RogerGinBer/RHermes")

Setup

RHermes can perform almost all its functions after installation, but the SOI Blank Substraction step requires a valid keras and tensorflow installation (which rely on Python).

Option 1: Default installation

In principle both keras and tensorflow can be configured with:

reticulate::install_miniconda()
keras::install_keras()
tensorflow::install_tensorflow()

After which you can check the following:

tensorflow::tf_config()
model <- keras::load_model_hdf5(system.file("extdata", "ImprovedModel.h5", package = "RHermes"))
is(model, "python.builtin.object") #Gives TRUE if the loading is successful.

If both commands don’t give any error (the “Your CPU supports …” warning is fine) the installation has been successful. If it fails (which can happen in some devices, try Option 2).

Option 2: Manual installation

First install Miniconda:

reticulate::install_miniconda()

Now find the Miniconda Prompt in your computer. Instead of relying on the default r-reticulate environment, type the following to create a new environment:

conda create -n newenv python=3.6 tensorflow keras

When finished, type in R:

reticulate::use_condaenv("newenv", required = TRUE)
tensorflow::tf_config()
model <- keras::load_model_hdf5(system.file("extdata", "ImprovedModel.h5", package = "RHermes"))

Everything should run smoothly. If not, try manually installing Anaconda from their website and letting reticulate know where to find the environment.

Also check out Keras and Tensorflow R tutorials.

Analyzing LC-MS data with RHermes

Once installed, you can use RHermes programmatically like this:

library(RHermes)
#Generate a Exp object
example <- RHermesExp()

#Set your formula and adduct database
example <- setDB(example, db = "hmdb")

#Process your MS1 files
example <- processMS1(example,
                        system.file("extdata", "MS1TestData.mzML",
                        package = "RHermes"))
#Generate SOIs
example <- findSOI(example, getSOIpar(), 1)

#Generate an IL 
example <- generateIL(example, 1, ILParam())

Or start the interactive GUI typing:

Please check the User Guide vignette for more detailed info and real examples.

Bug reporting

Suggestions and bug reports are more than welcome at: https://github.com/RogerGinBer/RHermes/issues

Citation

Please cite this package as:

HERMES: a molecular formula-oriented method to target the metabolome Roger Giné, Jordi Capellades, Josep M. Badia, Dennis Vughs, Michaela Schwaiger-Haber, Maria Vinaixa, Andrea M. Brunner, Gary J. Patti, Oscar Yanes bioRxiv 2021.03.08.434466; doi: https://doi.org/10.1101/2021.03.08.434466