Petri nets

ProCope contains methods to integrate information from an arbitrary of protein interaction networks, complex sets and purification data sets into a single colored Petri net. Below you find a description of the relationships contained in these Petri nets as well as a detailed specification of the different types of places and transitions participating in the nets.

Information integrated in the Petri nets
Creation of Petri nets and exporting to other formats
Types of places and transitions
Internal Petri net format

Information integrated in the Petri nets

Petri nets generated by ProCope contain different kinds of relationships from the networks, complex sets and purification datasets they are generated from:

all involved proteins along with eventual annotations for each protein
the binary interactions (edges) of the interaction network, labelled with the weight of that association and all further annotations for that edge
each complex along with the information of which proteins it consists
each purification experiment with its bait protein and prey proteins
consistency information:

binary interactions which are consistent with complexes (both proteins of the interaction are contained in the complex)
binary interactions which are consistent with a purification experiment
complexes which are consistent with single purifications (all proteins of the complex are contained in the purification)
purifications which are consistent with single complexes

Creation of Petri nets and exporting to other formats

The Petri net generation method primarily creates a file containing an internal text representation of the generated network (see below). This network file can then be converted to different formats. Currently two export formats are implemented:

XGMML - The Petri net is converted to the XGMML (eXtensible Graph Markup and Modeling Language) format. XGMML files can be imported into Cytoscape for example.
ToPNeT - A pair of ToPNeT compatible .places/.interactions files is generated.

Types of places and transitions

Below is a table of all different kinds of places you find in the Petri nets generated by ProCope. Each place has an attribute "type" which contains this information. In addition, each place has a "name" attribute which is used to label it e.g. for visualization. A unique internal ID is assigned to every place to identify it in the transitions.

Notes:
[n] denotes an automatically increasing integer number, [setname] is the name of the corresponding complex set or purification dataset.

Places type	Internal ID	"type" attribute	Other attributes
Protein	identifier of the protein	protein	"name" - same as internal ID
Binary interaction	bin_[n]	binary_interaction	"name" - same as internal ID
Complex	complex_[setname]_[n]	complex	"name" - same as internal ID
Purification experiment	purification_[setname]_[n]	purification_experiment	"name" - same as internal ID

Next the table of transitions, each transition has a "type" attribut and specific types of input and output places.

Notes: [network] is the name of the network a binary interaction comes from, [complexset] is the name of the protein complex set the complex comes from and [dataset] is the name of the purification dataset in which the respective experiment is contained.

Transition type	Input places	Output places	"type" attribute	Other attributes
"binary interact"	Proteins	Binary interaction	binary_interact	source = [network]
"consists of"	Proteins	Complexes	consists_of	source = [complexset]
"purfication"	Proteins	Purification experiment	purification	source = [dataset]
"consists of"	Binary interaction Binary interaction Complex Purification experiment	Complex Purification experiment Purification experiment Complex	consistent_with	-

The input arc (an edge in a Petri net) which leads from the bait protein to the "purification" transition is labelled with the string "bait".

Internal Petri net format

In the internal Petri net format produced by ProCope, each line represents one place or transition.

Places

A place line has the format:

p<tab>[ID]<tab>[annotations]



where

[ID] is a unique identifier of that place
[annotations] are optional key/value pairs in the format key1=value1;key2=value2...

note that values are optional, you can also write key1;key2=value;key3;key4

Example place lines:

p       ykl099c name=ykl099c;type=protein

p       ybl074c name=ybl074c;type=protein

Transitions

A transition line has the format:

t<tab>[in]<tab>[out]<tab>[annotations]

where

[in] and [out] are comma-separated lists of place identifiers
[annotations] are optional key/value pairs (see above)

Example transition lines:

t       bin31  
complex_MIPS_restricted_8      
type=consistent_with

t       bin103  complex_MIPS_restricted_23      type=consistent_with

Each incoming edge can also be labeled with annotations like this:

t    transition   
inplace1,(inplace2:key1=value1;key2=value2)   
outplace1,outplace2

ProCope documentation