View on GitHub

DLipid

Network inference and dynamic simulation application

Download this project as a .zip file Download this project as a tar.gz file

Welcome to DLipid Page.

DLipid is a Java and Matlab application for lipid flux analysis and dynamic simulations. The application uses time-course mass spectrometry data to automatically construct remodeling network and infer dynamic system using B-Spline algorithm.

DLipid is composed of two well-defined sub applications: FluxAnalysis and DynamicSimulation.

How to use it?

This application can be used in linux, mac, and windows.

To use this tool, you must first download and unpack DLipid.

$ tar -zxvf DLipid.tar.gz
$ unzip DLipid.zip

You will also need to install Matlab.

Step 1: Construct correlation network and simulation

$ cd DLipid/bin
$ java -jar DLipid.jar <inputFile> <minTh> [-m] [-n]

Usage:

$ java -jar DLipid.jar ../DATA/18_0-18_1.txt 0.3

A) inputFile:

Notice: The default input folder is "../DATA/". If only a filename is given, the application interprets it as "../DATA/filename".

Specificiation

                      time1           time2           . . .

      specie1         mean    std     mean    std

      specie2         mean    std     mean    std

       . . .

B) minTh : -- Significance score threshold for FluxNetwork application. -- Must be positive.

C) OPTIONS:

-mFLOAT : Maximum threshold. If this value is specified, multiple runs of simulations will be performed using thresholds from minTh to this number at step size 0.1.
-nINT: Iteration number in B-Spline algorithm, default is 1000. The algorithm will stop at next-closest iteration where the error is min within last four iterations. The algorithm will also stop when error change is less than 0.0001 between subsequent steps.

Usage:

$ java -jar DLipid.jar ../DATA/18_0-18_1.txt 0.3 -m0.5 -n1500

Then you will see Matlab application starts after a few seconds. You can quit Matlab after calculations finish.

Step 2: Retrieve correlation network and simulation results.

Go to output folder.

If only a filename is given in Step 1, default output folder is "../RESULTS/$inputFile_prefix/". Otherwise an output folder "$inputFilePath/$inputFile_prefix/" will be created.

If you excite the first example command at Step 1, you will see the following files in the output folder.

"log_18_0-18_10.3.txt": stores parameter values and information from dynamic simulations.

"sim_18_0-18_10.3.jpg": shows the fit of simulation to observed kinetics.

"18_0-18_10.1.graphml": is a xml format file of inferred flux network. See Graphml File Extension for more detail.

"18_0-18_10.3.sif": is a sif format file of reaction network. Specification:

source1 parameter1 target1
source2 parameter2 target2
...

LipidGraphical: A graphical interface for flux analysis.

This application allows vitalization of .grahpml network files.

$ java -jar FluxAnalysisGUI.jar

DynamicSimulation:

This application allows simulation with user-defined .sif reaction network file.

$ cd bin
$ matlab

Usage:

> main_remodeling_sim('../DATA/18_0-18_1.txt',
                     '../RESULTS/18_0-18_1/18_0-18_10.3.sif');

Authors

Norberto Díaz-Díaz (ndiaz@upo.es) and Lu Zhang (zhanglv@bc.edu).