A Platform-Based Design Environment for Synthetic
Biological Systems
Douglas Densmore1∗, Anne Van Devender1+, Matthew Johnson2∗, Nade
Sritanyaratana2∗
Department of Electrical Engineering and Computer Sciences1
Department of Bioengineering2
University of California, Berkeley∗
Washington and Lee University+
vandevendera@wlu.edu {densmore@eecs, matthewjohnson, nadesri}.berkeley.edu
ABSTRACT
Genomics has reached the stage at which the amount of
DNA sequence information in existing databases is quite
large. Synthetic biology is now using these databases to cat-
alog sequences according to their functionality thus creating
a system of standard biological parts. Flexible tools are
needed which both permit access and modification to that
data and also allow one to perform meaningful, intelligent
manipulation. A Platform-Based Design approach views ge-
netic information as having a particular functionality and
assembles platforms (collections of DNA elements) to per-
form this functionality. Specifically this paper presents the
Clotho toolset which uses these concepts to create a com-
plete design environment for standardized biological parts.
Categories and Subject Descriptors
D.2.2 [Software Engineering]: Design Tools and Tech-
niques
General Terms
Design, Management
Keywords
Platform-based Design, Synthetic Biology
1. INTRODUCTION
Synthetic biology is a rapidly growing field in which the tech-
niques of chemistry, biology, and engineering merge. Syn-
thetic biologists look to create new microorganisms by ma-
nipulating the basic building blocks of life to create living
material which interacts with, manipulates, and responds to
the environment in which it lives. Synthetic biology is very
much a design science where a new system is created by re-
searchers in laboratories using a series of design steps along
with their understanding of biological processes. Synthetic
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. To copy otherwise, or
republish, to post on servers or to redistribute to lists,requires prior specific
permission and/or a fee.
Tapia’09, April 1-4, 2009, Portland, Oregon, USA.
Copyright 2009 ACM 978-1-60558-217-7/09/04...$5.00
biology has the potential to greatly impact greater society
through the development of new technologies in drug pro-
duction, biofuels, and drug delivery vessels.
In an attempt to standardize this process, leverage previ-
ous design experiences, and begin to create a predictive de-
sign environment, registries of standard biological parts are
beginning to emerge [14]. Researchers have begun to talk
about how to classify these parts, create CAD systems, and
establish standards [13], [12], [7], [19], [10]. [18] lays out
very nicely an example of how these parts can be used to
program bacteria and discusses how they can be character-
ized (e.g. sensors, switch logic, inducers, etc). The fact
that these collections of parts can be discussed in terms of
their functionality along with rules for their composition,
raises the interesting question of how the Electronic Design
Automation (EDA) community (traditionally in electrical
engineering and computer science) possibly can leverage its
techniques in the creation of biological systems.
This paper describes the design of a toolset called Clotho
(named after the Greek fate which spun the thread of life)
which uses a methodology called Platform-Based Design (PBD)
[16] to approach the problem of designing synthetic biolog-
ical systems. In particular, we will describe its separation
of computation, communication, and coordination, the con-
cept of a “platform” as a common semantic meeting place
for designs, and the notion of both “top down” and “bottom
up” design styles.
1.1 Requirements
In the world of biology one can roughly separate tool offer-
ings into three broad categories. The first category are those
tools which provide computational power to specific biologi-
cal algorithms. BLAST (Basic Local Alignment and Search
Tool) [15] aligns nucleotide and protein sequences to allow
for functional prediction and to aid in locating sequences in
databases. ORBIT [11], [4] is protein design software which
allows the design of an amino acid sequence that folds into a
particular 3D structure. Mfold [20] enables the prediction of
mRNA secondary structures which aids in predicting mRNA
regulation and ribosome binding site strengths. These types
of tools require a strong understanding not only of the un-
derlying biology but also that the biology be predictive.
The second category of tools are those tools which allow the
24

user to design biological systems. Many biological systems
are not well understood, thus requiring a very empirical ap-
proach in which one must observe a number of experiments
and create hypotheses based on the collected data. In order
to perform this research in an efficient, repeatable manner
tools must exist to help in this process. Examples of these
tools are APE (A Plasmid Editor) [5], BioJADE [8], Gene
Designer [17] by DNA 2.0, and GenoCAD [3].
Finally, the third category are “glue tools” which do not
require a great deal of biological knowledge but are used
routinely by biologists in a laboratory setting. Function-
ality needed includes: translation of a gene into an amino
acid sequence, producing the reverse complement of a se-
quence, calculating the melting point for a region of DNA,
identifying restriction sites and other DNA motifs, calcu-
lating transmembrane regions, calculating the probability
that a protein has a secretion signal sequence (signalP), etc.
This list continues to grow and each lab has its own set of
favorites.
Mapping View
(Assignment of functionality to parts)
Part
s Vi
ew
(Com
pon
ents
for
a de
sign
)
Functional View(Behavior for a design; Part agnostic)
RBS Regulatory
1. Multiple Views of the Design
Signalling Protein Gen.ACGTACGGTT
View API 1 View AP2 View API N
Functionality 1(Codon Optimize) Functionality 2(Sequence Align)
Functionality 3(mRNA analysis) Functionality N
User Interface(Command Line, Window Based, GUI, Scripts, etc)
Menu MenuTitle
Button Tool Bar
Data
base

API
1
(Rel
atio
nal)
Data
base

API
2 (X
ML)
Data
base

API
N
XML
MySQL
Relational
Too
l AP
I 1
(Com
puta
tion
)
Too
l AP
I N
(Sim
ulat
ion) Input/Output, Import, Export Support(FASTA, GenBank, Proprietary Formats)
2. Interface to Part Repositories
3. Interface to Computational Tools
5. Interface to Simulation Tools
i.e. BLAST
i.e. Matlab
4. Push Info to the User
Figure 1: Framework for the Design of Synthetic
Biological Systems
This work recognizes the fact that a successful design envi-
ronment will encompass all three categories. It should do so
in such a way that allows for the tool to grow and expand
while not being overwhelming complex for an end user. Any
new tool offering in the design area needs to minimally sup-
port five aspects.
1. Provide various views of the design. It is im-
portant that the designer be able to view the system
at various abstraction levels and from various perspec-
tives. Examples of abstraction levels in DNA manipu-
lation may include viewing the entire genome, individ-
ual genes, and various nucleotide sequences. Perspec-
tives include DNA“design” functionality (e.g. sensors,
actuators, switches) or DNA “biology” functionality
(e.g. terminators, ribosome binding sites, promoters).
2. Interface with both local and remote part repos-
itories. The tools should support both importing and
exporting parts to well know databases such as MIT’s
Registry of Standard Biological Parts [14].
3. Support the interaction with computational tools.
The environment should allow the design under exam-
ination to be operated on by a large variety of external
computational tools. It should also support the export
and import of designs in a variety of standardized for-
mats (such as FASTA, GFF3, and GenBank).
4. “Push” information to the user. The tools should
not passively allow for the designer to create a sys-
tem. Constant feedback as to the validity of the design
should be provided whenever possible.
5. Provide simulation support. Simulation of a com-
pleted system or aspects of the system should be avail-
able when such simulation engines exist [2]. The in-
ternal design representation should be modular as to
allow subsections to be extracted for this purpose.
Figure 1 illustrates such a framework with these five aspects.
Platform-based design will very nicely lend itself to the first
requirement (various views) as will be shown. The other
aspects will require a very structured software engineering
approach with a polymorphic, inheritance based approach
to API design. These issues are touched on in Section 2.
Finally we should state the goals of such a design envi-
ronment clearly. A tool in this space should take the de-
sign of small biological circuits and systems (<20,000 base-
pairs) and push it to the whole engineering of a genome (4
megabases+). It should allow both designs for commercial
synthesis as well as design in by researchers in an academic
setting (i.e. be technology agnostic). It should also be freely
available under an open source ideology such as BSD.
1.2 Organization
The rest of this paper is organized as follows: Section 2 dis-
cusses Clotho’s system architecture. This demonstrates how
we adhere to PBD concepts as well as those of disciplined,
object oriented software design. Section 3 discusses Clotho’s
design flow. We illustrate how each of the five aspects of a
successful design tool are included. We also show how “top
down” and “bottom up” design can be achieved. Finally,
Section 4 provides conclusions and future work.
2. CLOTHO SYSTEM ARCHITECTURE
Platform-based design is a very complex methodology which
cannot be completely covered in this paper. For brevity’s
sake we will focus on three aspects of PBD in this paper:
25