This repository contains the tools that will be ported into the iPlant project. As a starting point, the tools consist of:
-
GP2S: A robust Bayesian two-sample test for detecting intervals of differential gene expression in microarray time series. doi:10.1089/cmb.2009.0175
-
CSI: Causal structure identification algorithm for inferring gene regulatory networks. How to infer gene networks from expression profiles, revisited. doi:10.1098/rsfs.2011.0053
-
HCSI: Hierarchical variant of the CSI. Nonparametric Bayesian inference for perturbed and orthologous gene regulatory networks. doi:10.1093/bioinformatics/bts222
-
Wigwams: identifying gene modules co-regulated across multiple biological conditions. doi:10.1093/bioinformatics/btt728
-
Apples: Analysis of Plant Promoter-Linked Elements. http://www2.warwick.ac.uk/fac/sci/systemsbiology/staff/ott/tools_and_software/apples/
-
Gradient Tool: not sure where this is from. Looks for significant changes in gradient of timeseries. http://www.plantcell.org/content/23/3/873.long
-
VBSSM: A Bayesian approach to reconstructing genetic regulatory networks with hidden factors. doi:10.1093/bioinformatics/bti014
-
MVBSSM: not sure how this variant alters the above.
-
MemeLab: motif analysis in clusters. doi:10.1093/bioinformatics/btt248
-
Wellington: a novel method for the accurate identification of digital genomic footprints from DNase-seq data. doi:10.1093/nar/gkt850
- Dafyd's "Switch Tool" (AKA Retros, currently Matlab) (from PRESTA Project)
** Regulatory Switch Model/Tool, with Giorgos
Some of these projects are already written in languages compatible with the iPlant infrastructure, but others are based in Matlab and will be ported to Python and C++ for compatibility and performance.
-
Krzysztof suggested various outputs for CSI. Cytoscape is a commonly used program for displaying network graphs, SIF and EDA formats would be useful. SIF defines the graph, including nodes and edges, EDA defines attributes on the edges---which is where I presume the weights would go?
-
Krzysztof also agreed that the web page output of CSI could be very useful. Users would want to be able to see the expression profiles for selected genes (one, not sure if any more?) and be able to see the graph truncated at specific weights. Would then want to be able to save/export out the data for e.g. Cytoscape. Javascript Service Workers should make this sort of saving posible?
Tools could have a header at the beginning of their main report/PDF containing the name of the creating tool along with citation information.
The file system used within iPlant is called iRODS. It comes with
its own set of iCommands for performing ftp
like operations at the
command line. Alternatively there are GUI apps like iDrop and a
fuse
plugin available. There are also nice Python libraries
available for talking to iRODS servers, for example python-irodsclient.
To connect to the iPlant servers, you need the following details:
host: data.iplantcollaborative.org
port: 1247
zone: iplant
username: <iplant username>
password: <iplant password>
These are the minimal Python libraries that would need to be installed:
pip install numpy scipy pandas h5py matplotlib GPy
In order to run an interactive exploration in iPython, these are also needed [google]:
pip install ipython pyzmq jinja2 tornado jsonschema