E-Cell System Version 3 Features

Basic capabilities

• Object-oriented modeling and simulation of complex systems.
• Plug-in architecture. New user-object classes can be developed, dynamically loaded, and used in simulation.
• Real-time user interaction and visualization during the simulation.

Simulation algorithms

(Simulation methods below can be mixed in any combination in a model.)

Algorithm	Corresponding modules	Maintainer	Status
Gillespie-family Stochastic Simulation Algorithms
Gillespie-Gibson	GillespieProcess/NRStepper	shafi/kaizu	S
Explicit/implicit Tau-leap	TauLeapProcess/Stepper	kaizu/tomo	B
Langevin method		(kaizu)	D
Ordinary Differential Equations
Radau5/Dormand-Prince adaptive	ODEStepper	kaizu	S
Dormand-Prince 4(5)7M explicit	ODE45Stepper	kaizu	O
Fehlberg 2(3) explicit	ODE23Stepper	kaizu	O
Euler explicit	FixedODE1Stepper	kaizu	O
Differential-Algebraic Equations
Radau5 implicit	DAEStepper	kaizu	S
Power-law formalisms
S-System	SSystemProcess/Stepper	(kmaru)	DU
Generalized Mass-Action	GMAProcess/Stepper	(kmaru)	DU
Other simulation methods
Flux distribution hybrid		tomo	S
Discrete-time simulator	DiscreteTimeStepper	shafi	S
Generic discrete event	DiscreteEventStepper	shafi	B

Scripting

The scripting interface is provided in Python language.

Feature	Corresponding modules	Maintainer	Status
Simulation session scripting	ecell3-session command	shafi	S
Analysis session scripting	ecell3-session-manager command	sugi	S
Model file generation	ecell.emllib Python module	zak	S

Compatibilities

Feature	Corresponding modules	Maintainer	Status
SBML level 1/2 importing	ecell3-sbml2eml command	(ishida)	SU
SBML level 1/2 exporting	ecell3-eml2sbml command	(ishida)	SU

Parallel computation

Feature	Corresponding modules	Maintainer	Status
Parallel Stepper scheduling	libecs/Scheduler	satya	D
Distributed computing	ecell3-session-manager command	msugi	S

---

Definitions:

Simulation session is..

A single run of a model under certain parameter set / condition. Also available as ecell.Session Python module.

Analysis session is..

A simulation experiment that involves many runs of simulation sessions (such as parameter tuning, metabolic control analysis etc..). Also available as ecell.SessionManager Python module.

Model file generation is..

Automated model generation, e.g. from databases.

Parallel Stepper scheduling is..

A shared-memory, multi-thread parallelization of a single simulation session by stepping multiple Steppers simultaneously, when possible. (to be merged into the main branch.)

Distributed computing is..

Running multiple simulation sessions on different computation nodes on cluster or grid environments. Currently Sun Grid Engine and Globus toolkit are supported.

---

Status::

S -- Stable. B -- Beta. D -- Under development. O -- Old (Deprecated; possibly removed from package in future releases). U -- Currently unmaintained, but intended to be part of the feature set.