the threshold parameter (in which case the interaction is functional)
and below the threshold parameter (in which case the interaction is
nonfunctional). For a detailed description of how this is achieved,
see Huang et al. [60].
In the ensemble generated by RACIPE, all models have the
same topology but differ in the values of kinetic parameters governing the model dynamics. The dynamics of each model is then
numerically simulated multiple times, each time starting with a
different set of initial concentrations of the molecules in the circuit.
This allows RACIPE to obtain a set of steady states that a given
model can generate. Once this has been done for each model in the
ensemble, RACIPE obtains a collection of steady states that the
given circuit topology can exhibit. Each model in the ensemble
generated by RACIPE may be interpreted as representing a single
cell. Thus, the collection of steady states obtained by RACIPE will
represent an in silico gene expression profile obtained for a population of cells. One aspect that should be kept in mind is that a model
that can exhibit more than one steady states will be counted more
often in the collection of steady states generated by RACIPE as
compared to a model that can exhibit only one steady state. Nevertheless, this steady state expression data can be analyzed using
familiar methodologies including principal component analysis
and hierarchical clustering to gain insight into the different classes
of steady states that may be exhibited by a given network topology.
The C language computer code implementing the RACIPE
framework is available online on GitHub (https://github.com/
simonhb1990/RACIPE-1.0). Once the code has been downloaded, change to the folder or directory where the code files are
present and use the make command to compile the code files for
your system. This will generate a single executable named
“RACIPE.” This executable takes as input a topology file, extension .topo, which describes the topology of the circuit being analyzed. This must be a plain text file with three tab-separated
columns. The first column (“Source”) contains the name of the
regulator gene. The second column (“Target”) contains the name
of the gene being regulated. The third and final column (“Type”)
describes the interaction type, 1 if the interaction is activating and
2 if the interaction in inhibiting. A sample topology file (TS.topo) is
available online with the code. Once the topology file for the circuit
of interest has been generated, the RACIPE code can be run as
follows:
$ ./RACIPE network.topo
Additional input options that may be provided to the code are
described in the “README.md” file available with the code. Upon
execution, the code generates multiple files. Most important
among these are:
Mathematical Modelling of EMT
395
and below the threshold parameter (in which case the interaction is
nonfunctional). For a detailed description of how this is achieved,
see Huang et al. [60].
In the ensemble generated by RACIPE, all models have the
same topology but differ in the values of kinetic parameters governing the model dynamics. The dynamics of each model is then
numerically simulated multiple times, each time starting with a
different set of initial concentrations of the molecules in the circuit.
This allows RACIPE to obtain a set of steady states that a given
model can generate. Once this has been done for each model in the
ensemble, RACIPE obtains a collection of steady states that the
given circuit topology can exhibit. Each model in the ensemble
generated by RACIPE may be interpreted as representing a single
cell. Thus, the collection of steady states obtained by RACIPE will
represent an in silico gene expression profile obtained for a population of cells. One aspect that should be kept in mind is that a model
that can exhibit more than one steady states will be counted more
often in the collection of steady states generated by RACIPE as
compared to a model that can exhibit only one steady state. Nevertheless, this steady state expression data can be analyzed using
familiar methodologies including principal component analysis
and hierarchical clustering to gain insight into the different classes
of steady states that may be exhibited by a given network topology.
The C language computer code implementing the RACIPE
framework is available online on GitHub (https://github.com/
simonhb1990/RACIPE-1.0). Once the code has been downloaded, change to the folder or directory where the code files are
present and use the make command to compile the code files for
your system. This will generate a single executable named
“RACIPE.” This executable takes as input a topology file, extension .topo, which describes the topology of the circuit being analyzed. This must be a plain text file with three tab-separated
columns. The first column (“Source”) contains the name of the
regulator gene. The second column (“Target”) contains the name
of the gene being regulated. The third and final column (“Type”)
describes the interaction type, 1 if the interaction is activating and
2 if the interaction in inhibiting. A sample topology file (TS.topo) is
available online with the code. Once the topology file for the circuit
of interest has been generated, the RACIPE code can be run as
follows:
$ ./RACIPE network.topo
Additional input options that may be provided to the code are
described in the “README.md” file available with the code. Upon
execution, the code generates multiple files. Most important
among these are:
Mathematical Modelling of EMT
395