Biomedical Engineering Reference
In-Depth Information
assembly (i.e., for full combinatorial reuse of each DNA input
fragment), each scar-less assembly junction sequence must be
found within a region of sequence identity across all combinato-
rial variants to be constructed.
The option to choose between multiple putative scar-less
assembly junction sequences presents optimization opportunities as
well as challenges. There are a multitude of sensible assembly junc-
tion sequence optimization strategies, including but not limited to:
avoiding sequences previously resulting in poor assembly effi cien-
cies, avoiding sequences with strong single-stranded DNA second-
ary structure (e.g., hairpins), optimizing sequence annealing
temperature and GC content, avoiding sequences that are signifi -
cantly homologous to other assembly junction sequences or other
sequences found within the DNA to be assembled, achieving 3
′
GC-clamps where possible, minimizing sequence length and cen-
tering the sequence location between the corresponding pair of
DNA sequences to be assembled (to minimize input costs).
Furthermore, any given optimization strategy may not be equally
effective for all DNA assembly methodologies (e.g., for CPEC,
avoiding assembly junction sequences that are signifi cantly homolo-
gous to other sequences found within the DNA to be assembled is
likely to be important, but this may not be of concern for the
Gibson method [
2
]). The challenges introduced by the option to
choose from a set of putative assembly junction sequences, then, are
twofold: the additional burden for each new construct of having to
select an optimal sequence for each scar-less assembly junction, and
the necessity to decide how to prioritize and triage the various opti-
mization strategies for a given DNA assembly methodology.
Design automation software can relieve the burden of select-
ing and optimizing new scar-less assembly junction sequences, and
can predict and potentially mitigate poor assembly junction
sequence performance (e.g., via hierarchical DNA assembly
design). Several software tools have been developed to help auto-
mate the design of protocols for each of the three categories of
scar-less DNA assembly methods presented in Table
1
. It is very
important to point out, however, that the various software tools
vary wildly in terms of their sophistication, functionality, fl exibility,
and especially in terms of their respective capacities to optimize
assembly junction sequences and predict and mitigate poor assem-
bly junction sequence performance. As it is beyond the scope of
this chapter to review the functionalities of the various software
tools presented in Table
1
, it is merely suggested here that it is very
worthwhile to briefl y evaluate each of the tools in turn to deter-
mine which is the most appropriate or useful for a given set of
DNA assembly protocol design tasks. This is especially prudent,
since software tools that appear to be useful on paper are often
confounded by factors (as mundane as software inaccessibility, slow
web-server response times, or poorly crafted user-interfaces) that
limit their suitability in practice.
