epaull / tiedie Goto Github PK

View Code? Open in Web Editor NEW

36.0 10.0 22.0 97.7 MB

Tied Diffusion for Subnetwork Discovery (TieDIE)

Home Page: http://bioinformatics.oxfordjournals.org/content/early/2013/08/27/bioinformatics.btt471.abstract

License: Other

Shell 0.52% MATLAB 3.75% R 12.47% Python 83.27%

tiedie's Introduction

TieDIE: Tied Diffusion for Subnetwork Discovery.

Current Version

1.0

Authors

Evan O. Paull, Daniel Carlin and Joshua M. Stuart.

Additional Contributors

Srikanth Bezawada (TieDIE Cytoscape Plugin) Josh L. Espinoza (Quick kernel loading feature) Dana Silverbush (MATLAB kernel generation code updates to newer versions)

Requirements

Python 2.7 and the python numpy module are required to run the tiedie executable, when using a pre-computed diffusion kernel.

Either MATLAB or the python scipy module, version 0.12 or later, is required for diffusion kernel computation: the latter is free, though not as computationally efficient as the MATLAB implementation.

python 2.7: all modules.
- scipy: >= 0.12.0 kernel generation
- numpy
- networkx
MATLAB: kernel generation

Installation

Install dependencies
Download the TieDIE repository to the desired location
(Recommended) Pre-Generate kernel file with MATLAB (bin/makeKernel.sh)
Run TieDIE/bin/tiedie

Examples

GBM.test An example signaling network from Glioblastoma (TCGA Network, 2012) is provided, along with input heats for an upstream set of genes (mutated genes) and a downstream set of nodes (transcriptional responses). To run:

cd examples/GBM.test make

A tutorial for this simple example is provided in doc/Tutorial.pdf.

Programs

tiedie Python executable to run the TieDIE algorithm.
makeKernel.sh Shell script executable that calls MATLAB for diffusion kernel file generation.
(Auxillary) span.R An R-implementation of the Prize Collecting Steiner Tree network formulation, that calls the BioNet package.

Folders

bin : executables and matlab source files
lib : python code libraries for the tiedie executable
test : doctest unit tests, functional tests and regression tests
examples : GBM and BRCA inputs for demonstration purposes
galaxy : Galaxy web-server wrapper for tiedie to run through the web interface. (https://main.g2.bx.psu.edu/)
pathways : the "superpathway" described in the TieDIE paper, used with the TCGA BRCA dataset

In Press

TieDIE was first featured in the 2013 Nature paper "Comprehensive molecular characterization of clear cell renal cell carcinoma". In this TCGA (The Cancer Genome Atlas) network publication, a TieDIE analysis was used to connnect frequently mutated genes involving the SWI/SNF chromatin remodelling complex to a diverse set of gene expression changes characteristic of tumor development and progression. The TieDIE manuscript was not yet published at the time of the Nature publication and so is cited by name and author only. The TieDIE network solution is shown in figure 4 of the main text, which can be found at this link: http://www.nature.com/nature/journal/v499/n7456/full/nature12222.html .

Contact

Feature requests, comments and requests for clarification should all be sent to the author at [email protected]. I will try to respond quickly to all requests, so feel free to email me!

tiedie's People

Contributors

Stargazers

Watchers

tiedie's Issues

PPRdiffuser with no instance of attribute getLabels

In TieDie/examples/BRCA.Basal.vs.LuminalA, when I was trying to run the default example with pagerank option as

../../bin/tiedie -u upstream.input -d downstream.input -n tiedie-1.0-basal.vs.luminal.sif -s 1.0 --pagerank

and it prints into terminal as :

Traceback (most recent call last):
File "../../bin/tiedie", line 318, in
k_labels = diffuser.getLabels()
AttributeError: PPrDiffuser instance has no attribute 'getLabels'

The tiedie executable executes correctly without --pagerank options, but diffuser.getLabels() does not attribute any labels as k_labels.

How important is to check the node universe matches the k_labels for pagerank algorithm in tieDIE? Is this attribute supposed to be happen? When I commented out line 319 - 322 in ..bin/tiedie, it compiles and works fine.

Please let me know if you have any questions regarding to this issue, and thank you in advance

TieDie requires heats for upstream and downstream, how do you suggest deriving these

As the title states, I'm trying to figure out how I should derive some heats/weights to be used with this neat tool of yours. Any suggestions? I have a raw network where I have calculated heat values for my upstream subset. I see in the GBM.test that you've put 100 next to the downstream entries. Is this standard practice?

Usage of command option -size

I have a question about the usage of -size commend option. In bin/tiedie code, it says that opts.size sets the desired relative size of the linker set of genes to be found by the algorithm. opts.size passed as size_control global variable and it only used in findLinkerCutoff in line 133 and line 402.

In def extractSubnetwork(up_heats, down_heats, up_heats_diffused, down_heats_diffused, size_control, set_alpha) in bin/tiedie line 106, it calls the findLinkerCutoff when the alpha value is none. findLinkerCutoff calls findLinkerCutoffMulti when there are two sets(source set, target set),and at findLInkerCutoffMulti line 411 tiedie_util.py.

The parameter size in findLInkerCutoffMulti only used to check if the size is 0 or not, and does not have any other use in the function, and filterLinekrs call (up_heat_diffused,down_heat_diffused,1) with the set size of 1 for filtering the minimum score at 411 tiedie_util.py.

I am having hard time understanding the proper size of cutoff for large network, and would you explain what would be appropriate way to find the cutoff? I have tried setting different sizes with using pagerank, but they all resulting the entire graph in tiedie.sif file.

Please let me know if you have any questions, and thanks in advance.

can I use TieDIE applied to analyze different level of proteome, such as proteome and phosphoproteome

Feature Request | Python 2 & 3 Compatibility

Are there any plans to port this to Python 3 and then make it backwards compatible? If not, I can maybe take a crack at it. Are there any dependencies on out-dated modules?