Information Technology Reference
In-Depth Information
2 Models in Biology
Models are used intensively by experimental biologists to describe their mecha-
nistic understanding of living processes, however until recently they were most
often informal pictures and diagrams representing interactions between different
components of a biological system, for example signalling between cells or inter-
actions between proteins within a cell. In recent years, new modeling languages
and tools have been introduced, with the aim of providing formal semantics to
diagrammatic languages that can be used by biologists to describe their systems
and then gain the benefits of executable models. To help deal with complex-
ity, methods and languages from the programming languages community have
been introduced and adapted for biological applications, utilizing process calculi,
statecharts and rule-based formalisms. Overall progress in the field is allowing
the construction of exciting models that are increasingly grounded on biologi-
cal knowledge, and so offering opportunities for predictive capabilities, together
with raising a more urgent need to systematically analyze runtime properties.
3 Model Refutation and Verification
In software and system engineering, a main challenge is to improve the con-
fidence that a system satisfies a given specification, where specifications can
be described using, e.g., automata, temporal logic, pre- and post-conditions. In
studying natural biological systems, the specification is unknown, so this can
be viewed as a reverse engineering problem, where the scientific process aims
to construct models and theories about how the system works and identify the
specification. A model should be able to explain and reproduce the experiments
and data, and should typically not produce runs that are in contradiction with
known experiments. Thus effective ways to compare a model against known ex-
perimental results and hypotheses is crucial to allow the scientific research. Par-
ticular challenges for biological models is that they are highly parallel, include
probabilistic decisions representing stochastic elements, and are often very large
and computationally intensive to simulate. The runtime verification community
can contribute to biological research, by designing new specification formalisms,
improving runtime verification methods, and targeting new biological areas in-
cluding recent developments in the ability to construct computational circuits
from biological material [2,3], applications to reprogramming cells for medical
applications, and global ecological models [1].
References
1. Purves, D., et al.: Ecosystems: Time to model all life on earth. Nature 493(7432),
295-297 (2013)
2. Qian, L., Winfree, E.: Scaling Up Digital Circuit Computation with DNA Strand
Displacement Cascades. Science 332(6034), 1196-1201 (2011)
3. Yordanov, B., Wintersteiger, C.M., Hamadi, Y., Kugler, H.: SMT-based Analysis
of Biological Computation. In: Brat, G., Rungta, N., Venet, A. (eds.) NFM 2013.
LNCS, vol. 7871, pp. 78-92. Springer, Heidelberg (2013)
 
 
Search WWH ::




Custom Search