Skip to main content

Workpackage 3

Development of a Systems Medicine approach to infer condensed context-specific pathways

PI: Frank Kramer

The Pathways work package, led by Frank Kramer and Tim Beissbarth, is developing a Systems Medicine approach to infer condensed context-specific pathways based on prior knowledge, high-throughput and clinical patient data.

This workpackage is comprised of the steps of (a) developing a data model for pathway knowledge exchange, (b) data integration and (c) pathway analysis and interference.

(a) Developing a data model for pathway knowledge exchange
Networks are powerful and flexible methodology for expressing biological knowledge for computation abd communication. Albeit its benefits, the sharing of networks, the collaboration on metwork curation, keeping track of changes between different network versions, and detecting different version itself, still is a major problem in network biology. 
We developed a software package which provides an implementation of the Cytoscape Cyberinfrastructure (CX) Format in R, representing the CX information in an RCX object as well as an extended igraph object RCXgraph

Link RCX

(b) Data integration
 Based on our work in (a) we implemented a generic R package to
interact with the Network Data Exchange (NDEx). NDEx, is an open-source
software framework to manipulate, store and exchange networks of various
types and formats (Pratt et al., 2015, Cell Systems 1, 302-305, October
28, 2015 ©2015 Elsevier Inc. ScienceDirect). NDEx can be used to upload,
share and publicly distribute networks, while providing an output in
formats, that can be used by plenty of other applications. Our tool
called ndexr has been successfully published (Zitat Auer F, Hammoud Z,
Ishkin A, Pratt D, Ideker T, Kramer F. ndexr - an R package to interface
with the Network Data Exchange. Bioinformatics. 2017.) and helped us
integrate data from different sources like NDEx as well as
BioPAX-encoded pathway databases.

(c) Pathway analysis and inference
Using deep-learning techniques we have used the integrated pathway
knowledge as well as graph-based convolutional neural network for novel
pathway analysis methods.