Hi. I'm trying to create a network that will present all the pathways found in the KEGG database and the relations between those pathways. I want to represent each pathway as node, and compounds that are shared by two (or more...) pathways as edges. I have not found a tool to perform this task yet. I would be happy to get your advice on this issue. Thanks!
This is a common question around here. You didn't give us much information to go on so it's hard to make a clear suggestion, but for my own KEGG data I'm quite happy using Cytoscape and also have recently been communicating networks using hive plots.
I have a followup question, I want to identify incoming and outgoing edges for eg. http://www.kegg.jp/kegg-bin/show_pathway?hsa00010 to Pentose phosphate pathway (PPP) edge 1 (outgoing) from PPP edge 3 (incoming) citrate cycle 1, 1 (out and incoming) proponate metabolism 1 (out) carbon fixation 1 (in) starch and sucrose 1,1 (in, out)
How can I extract this from .xml or from any file? I can access all kegg file.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
What's your input ? a database ? a flat file ? a json file ?
The global map of pathways is kinda what you want: http://www.genome.jp/kegg/pathway/map/map01100.html
What kind of analysis are you trying to do by generating this map? Are you trying to weigh the pathway nodes by how many genes map to the pathway so you can get a enrichment value for the various pathways?
Damian, thanks for this link. I've only used "third party" applications to pathway map, so this is a nice addition to the tool box.
Hello Pierre & Damian. Pierre - my initial input is a list of pathways that I wish to map to the network described above. I tried to fetch the related pathways for each of the pathways listed in the KEGG db and use this data to generate a csv file (to use as input for cytoscape). However, I found that information not complete in the KEGG website. Damian - I'm familiar with the global metabolism map. What I want to do is to display the same information, with two main differences: displaying each pathway as a single node (and not few pathways grouped to one edge) and representing each compound, shared by two pathways as edge. What I want to do is to show that the subset of pathways I'm studying compose some kind of module, and are all connected by a certain path length from a specific pathway. Thanks!
Here's a highly relevant paper: http://www.biomedcentral.com/1752-0509/1/29