Biology Reference
In-Depth Information
Although GUI is convenient and makes it easy to design genetic circuits, there are some
limitations in that the user is limited to only what is available in the software. The
TinkerCell tool allows users to create and analyze synthetic networks using third-party C
and Python programs with an extensive C and Python application programming interface
(API). As an open-source tool, it allows for the development and implementation of new
tools that can be used in TinkerCell for those who have a background in computer
programing (
Table 8.1
)
.
35
COMPUTATIONAL TOOLS FOR PATHWAY PREDICTION
Advancing beyond the level of local circuit design in synthetic biology, the engineering of
biological systems at the enzyme or pathway level have been investigated for metabolic
engineering.
1,2,12,38
42
Pathway engineering can occur in all forms, from mixing and
matching pathway components into an ideal host to the prediction and design of novel or
de novo pathways that have not been identified in biological systems. While engineering
existing pathways and combining them to create new ones result in the efficient production
of desired molecules, this has limited our options to enzymatic reactions that are currently
known. This has led to the development of tools focused on the discovery of de novo
pathways, partially or as a whole, consisting of nonnatural enzymes that can expand the
known database of metabolism. With an ever-growing database and knowledge of
bioinformatics and metabolic pathways, it is often easy to think that the easiest way to
engineer genes for strain design is to use the enzymes with known functions and well-
characterized properties. The approach of identifying de novo pathways of nonnatural
enzymes, defined as enzymes that have not yet been characterized and are not commonly
found in known organisms, employs a method that scans the realm of all known chemical
reactions rather than known enzymatic reactions. This allows for the discovery of chemical
reactions that can potentially occur in a cell but are overlooked because the enzyme that
would catalyze such a reaction has not been characterized.
149
Enzymes have evolved through millions of years to carry out specific metabolic reactions.
However, many of these enzymes are promiscuous and have broad substrate specificities.
In vitro characterization of an enzyme may not fully explain the enzyme
sbehavioror
function in vivo, thus logical steps can be taken to hypothesize that an enzyme can
possibly perform the same function with multiple similar substrates. Therefore,
heterologous expression of enzymes that have been modified to efficiently utilize alternate
substrates can possibly lead to the production of nonnatural metabolites and the
discovery of a new metabolic pathway.
'
To facilitate the discovery of de novo pathways and their enzymes, a retrosynthetic analysis
is employed. Furthermore, algorithms have been developed to automate and efficiently
perform the analysis and present potential biochemical reactions to explore. Although de
novo pathway construction for strain design is limited by the biology of enzymes, it also
opens new opportunities for finding nonnatural metabolite-forming pathways that can be
ultimately useful for metabolic engineering.
Algorithms and computational tools that have been developed to address possible
biochemical pathways explore a wide range of chemical properties, including but not
limited to the thermodynamic feasibility of metabolic reactions under physiological
conditions. Computational tools such as the University of Minnesota Pathway Prediction
System (UM-PPS), DESHARKKY, the web-based PathPred from KEGG, Biochemical Network
Integrated Computational Explorer (BNICE), and EnzMatcher are examples of the tools
available for developing strategies in designing novel pathways.
1,2,38,39,43
The combinatorial explosion resulting in the addition of rule sets is also addressed in the
publically available UM-PPS, in which certain restrictions such as aerobic likelihood, relative
