I wanted to start analysing genome-scale metabolic models. I am interested in metabolic modelling to study human diseases. At least as of now, my idea is to use R for the analysis. This post describes all the technical set-up that I am doing on my ubuntu 10.04 LTS machine in order to eventually start working with these models.
As I am interested in human metabolic network, my model of interest is the reconstructed human metabolic model, recon 2. I googled to see where I can download it from. It has a web site http://humanmetabolism.org/. But the model is available for download from the supplementary material of this paper. There was a zip file that contained many files which I interpret as the generic human model and cell specific models for a variety of cell types.
I first wanted to see the contents of these files with my eyes. I know that I could open these files with any text editor, but I also knew that there is something called SBML editor which is specifically designed for viewing/editing SBML files. So I was tempted to install it. That didn't work too well at the first shot. Probably I could have got it to work if I tweaked a bit, but I decided to look for a generic XML viewer which would do the same thing as SBML editor for any XML file. So, I searched for XML editors and chose to try TreeLine and XML copy editor. Both of them were available from ubuntu repositories. So, disregarding a failed attempt at installing XML copy editor that was downloaded from the sourceforge page due to missing dependency packages, finally I installed both these editors from ubuntu repositories as follows:
sudo apt-get install treeline
sudo apt-get install xmlcopyeditor
As I am interested in human metabolic network, my model of interest is the reconstructed human metabolic model, recon 2. I googled to see where I can download it from. It has a web site http://humanmetabolism.org/. But the model is available for download from the supplementary material of this paper. There was a zip file that contained many files which I interpret as the generic human model and cell specific models for a variety of cell types.
I first wanted to see the contents of these files with my eyes. I know that I could open these files with any text editor, but I also knew that there is something called SBML editor which is specifically designed for viewing/editing SBML files. So I was tempted to install it. That didn't work too well at the first shot. Probably I could have got it to work if I tweaked a bit, but I decided to look for a generic XML viewer which would do the same thing as SBML editor for any XML file. So, I searched for XML editors and chose to try TreeLine and XML copy editor. Both of them were available from ubuntu repositories. So, disregarding a failed attempt at installing XML copy editor that was downloaded from the sourceforge page due to missing dependency packages, finally I installed both these editors from ubuntu repositories as follows:
sudo apt-get install treeline
sudo apt-get install xmlcopyeditor
Both are good, and I especially liked TreeLine's ability to render hierarchical tree view of the XML document. Thus, I think this viewer is perhaps better than SBML editor because it is more general. It also allows easy editing of the document. Of course SBML editor may certainly be more useful if you specifically only work with SBML files because of its 'awareness' of SBML schema. But for my purposes, which is at the moment only viewing, TreeLine is perfect (even in future I assume my editing, if I do any would be very simple and TreeLine should be perfect for that too).
After this, I installed an R package sybil which allows performing constraint based modelling of genome-scale models. To do this, I used the following R command
install.packages("sybil")
After taking a quick look at the documentation of sybil, I understood that I needed another package called sybilSBML to be able to read the SBML models. I tried the following R command to install it.
install.packages("sybilSBML")
The installation failed due to a missing dependency package called rsbml, which I realised is a Bioconductor package as install.packages("rsbml") failed. So, I tried installing this package by issuing the following R commands
source("http://www.bioconductor.org/biocLite.R")
biocLite("rsbml")
This installation failed because libsbml libraries were not installed on my computer. For the rsbml package to work the libsbml should be installed by using --with-libsbml-include option during configuration. So, installing the libsbml from ready made binaries is not enough. Therefore, I downloaded the core plus packages source (tar.gz) file, uncompressed the archive and tried to build it as follows
./configure --with-libsbml-include --with-java --with-python
make
Here --with-libsbml-include is the only option that I really need right now, but I gave the other two options just in case I would need in future. It failed with two errors. I don't remember the first error message now, but I interpreted that it was caused due to missing python-dev package. So I installed it from the ubuntu reposities as
sudo apt-get install python-dev
The second error message was about missing dependency package called swig. To fix this, I first tried to install swig from ubuntu repositories by doing
sudo apt-get install swig
But make script of libsbml didn't succeed still because swig version installed from ubuntu repositories was something like 1.3.x while libsbml expected swig 2.0.0. So, I downloaded swig.2.0.9 from its website and tried building it. The configure script of swig failed with an error which said that mkoctfile was not installed. Google told that the missing library that was needed to be installed was octave-headers. I installed it from ubuntu repositories as follows
sudo apt-get install octave3.2-headersAfter this, swig installation went well with the following commands
./configure
make
make install
After this, libsbml installation, which was done as follows, was successful.
./configure --with-libsbml-include --with-java --with-python
make
make install
Also the last lines of the installation output said that I will need to (1) run ldconfig (2) add /usr/local/lib to LD_LIBRARY_PATH. Right now I have only run ldconfig and didn't do the second thing. I will do it if something fails. I also installed the R package "R interface for libSBML" that was available among the files from sourceforge libSBML download page. I don't know if this is needed though.
After this, installation of R packages rsbml and sybilSBML was successful. That completes the platform setup for today. Next step is to test if everything works or something else needs to be done.
Comments