[an error occurred while processing this directive]

Murphy Lab

Home
Information
People
Addresses
Cytometry Development Workshop
FCS API

Research
Projects
Publications
Software
Presentations
Flow Cytometry

Affiliations
Carnegie Mellon University
Computational Biology Department
Center for Bioimage Informatics
Biological Sciences Department
Biomedical Engineering Department
Machine Learning Department
MBIC

[an error occurred while processing this directive]

Internal

Information
Lab FAQ
Lab Research Topics

Services
TypIX

Web Pages
Web Page Template
Visitation Statistics
Important Links

Software and Documentation
Matlab Docs
Khoros Docs
TN-Image Docs
Michael Boland's thesis

Murphy Lab - Software

The data and software used for the following paper can be downloaded below:

T. Zhao and R.F. Murphy. (2007) Automated learning of generative models for subcellular location: Building blocks for systems biology. Cytometry 71A:978-990.

The data and code are contained in five .tgz files (gzipped tar archives):

genmodel_codeonly.tgz 7 MB
- Update An updated version of the code is now available for download. It contains the source code for the mex files "ml_texture", "ml_moments", "ml_hitmiss", "ml_Znl" and "ml_multout". It also contains mex versions for Linux x64, Linux x32 and Mac x64. To download the updated package, click here.
genmodel_rawdataonly.tgz 300 MB
genmodel_procdataonly.tgz 2.6 GB
genmodel_modelsonly.tgz 200 KB (a small subset of genmodel_procdataonly.tgz)
genmodel_synimagesonly.tgz 84 MB (only the synthesized images from genmodel_procdataonly.tgz)

If you would like to regenerate all results from scratch, download and untar just the code and raw data and run ml_gendataall.m - this will take over a week on a single 1.7 GHz processor. It will generate the files that are in genmodel_procdataonly.tgz. Figures and tables in the paper can then be generated by running paperresults.m

If you would just like to regenerate the figures from the processed results, download and untar the first three .tgz files and run paperresults.m

If you would just like to generate images from the trained models described in the paper, download just the first and last files (codeonly and modelsonly) and run ml_testgenimg. The file can be modified to specify which models you wish to generate images from. Many examples of generated images can be found here. If you would just like to view the synthesized images used in the paper, download just the synimagesonly file.

System requirements

The scripts were written for Matlab on Linux, version 7.0 or above. The statistics toolbox, image processing toolbox, splines toolbox and optimization toolbox for Matlab are also required.

Tools created by others that are included in the code distribution for convenience

hts_code
LIBSVM
matlabPyrTools (used by textureSynth)
Netlab
OpenTSTOOL (used by hts_code)
textureSynth

Last Updated: 16 Sep 2011