=================================
Steady state network experiments:
=================================

There are 50 grendel networks and 50 abiochem networks, the networks have 100
genes each. The abiochem networks were derived from the A-BIOCHEM CenturySF networks: 
	http://www.comp-sys-bio.org/AGN/Century/index.html
The grendel networks were produced using our software.  The only difference
between abiochem and grendel networks is the network topologly.  

	The abiochem networks are scale free and follow a powerlaw for both in and 
out degree distributions.  The grendel networks in and out degree distributions 
follow an exponential and powerlaw respectively, and were fit to reflect the 
degree distributions of the transcriptional network in S. cerevisiae.  The 
kinetics of each network type is equivalent to the original parameterization of 
the CenturySF networks.


Each network was simulated over 5 different experimental design scenarios:

 Diverse: 

			300 measurements from a genetically and phenotypically diverse population
			this population was modeled by varying the efficiency of transcription
			and mRNA degradation across samples independently for each gene

			Files: abiochem_diverse.tar.gz grendel_diverse.tar.gz

 Knockouts: 

		100 measurements, systematically knocking out each gene

		Files: abiochem_knockouts.tar.gz grendel_knockouts.tar.gz

 Overexpression: 
		
		100 measurements, systematically overexpressing each gene

		Files: abiochem_overexpress.tar.gz grendel_overexpress.tar.gz

 Knockouts + overexpression: 
		
		200 measurements, knocking  out and overexpressing each gene

		Files: abiochem_knockouts_overexpress.tar.gz grendel_knockouts_overexpress.tar.gz

 Knockouts + 2 overexpressions: 

		300 measurements, knocking out and overexpressing each gene at two levels

		Files: abiochem_knockouts_overexpress2.tar.gz grendel_knockouts_overexpress2.tar.gz


=================================
Time course network experiments:
=================================

This is a collection of 250 networks with topologies that were generated
to match the degree distributions of the transcriptional network in
S. cerevisciae.  Each network has 20 genes and 2 environmental stimulii
(signals).  These networks are designed to be simulated as a time course.
Over a 2000 minute (33.3 hr) time course each network will undergo 4 
condition shifts where each external signal is perturbed and restored 
to its original state.  In the first 2000 minutes the system is allowed to
reach steady state.  In the next 2000 minutes the condition shifts are
carried out.

There are two sets of networks, one with kinetic parameters chosen in an arbitrary 
fashion equivalent to the A-BIOCHEM networks (dynamic_arb_param.tar.gz). The second 
where kinetic parameters were chosen to match real kinetic parameters in S. 
cerevisciae acquired from published data from genome wide experiments i
(dynamic_tru_param.tar.gz)