Understanding metabolism is fundamental in biomedical and plant research and the identification and quantification of thousands of metabolites by mass spectrometry in modern metabolomics is a prerequisite for elucidating this area. However, the identification of metabolites is a major bottleneck in traditional approaches hampering advances. Here, we present a novel approach for the untargeted discovery of metabolite families offering a bird's eye view of metabolic regulation in comparative metabolomics. We implemented the presented methodology in the easy-to-use web application MetFamily to enable the analysis of comprehensive metabolomics studies for all researchers worldwide. MetFamily is available under http://msbi.ipb-halle.de/MetFamily/.

The image sneumann/metfamily-base contains all dependencies for the MetFamily web application.

Build via docker build -t sneumann/metfamily-base -f Dockerfile-base .

The image sneumann/metfamily is built on top and contains the actual MetFamily code and web application.

Build via docker build -t sneumann/metfamily .

The build of the metfamily-base image https://hub.docker.com/r/sneumann/metfamily-base is triggerd whenever in the master branch a tag basechange-<date> e.g. basechange-20190804 is specified.

To run the resulting container, start with docker run --rm -p 3838:3838 sneumann/metfamily:latest

and point your browser to http://localhost:3838/

If you want to develop and debug stuff, you can build a container on top of metfamily:latest that has an added rstudio server. First build using docker build -t metfamily-rstudio -f Dockerfile-rstudio . and then run via docker run -it --rm -p 8787:8787 metfamily-rstudio:latest. CAVEAT: the Dockerfile-rstudio specifies a fixed user/password combo of rstudio:rstudio. Do not use in Production !

You can also pass a local directory with checked out MetFamily git tree via the docker run -v argument.

After installing the MetFamily R package and its dependencies, you can

library(MetFamily)
runApp(system.file("MetFamily", package="MetFamily"))

At IPB we are running MetFamily inside a Kubernetes cluster. Please contact us for questions.