Biological Computation: How does nature compute?
(From "Biological Computation NSF Workshop",
J.Hickman, F.Darema, W.R.Adrion)
When viewed as information technologies, it is apparent biological systems
have an enormous capability as control systems for agility or regulation,
for pattern recognition, adaptability, information storage, sensor fusion
and other information-handling tasks of great interest to computer
scientists, computer engineers and anyone else interested in IT-related
research. In these and other domains, biology performs at levels many
orders of magnitude better than silicon-based systems. We believe that
research at the interface of biology and information technology may lead to
important new information systems (algorithms and software) and computer
technologies (hardware). The question is what and how can we learn and
understand from the biological systems, and how can we adopt them and adapt
them to develop these new computer technologies. The objective of this
short treatise is to define what we mean by Biological Computation in view
of other work in this area and then develop the idea to serve as a basis
for future discussion.
Biocomputation
has been used as a catch-all term for research at the
interface of biology and computation, but that term is used in so many ways
and for such different subsets of this intersection as to cause confusion.
To help guide discussions, we offer that the general area of biocomputation
can be divided into four major categories:
Biomolecular Computation,
Computational Biology, Bioinformatics, and
Biological Computation.
We define the four categories as follows:
-
Biomolecular Computation
This category includes efforts to
exploit biological macromolecules to implement relatively standard methods
of computation. Examples are DNA computing, storage media using bacteria
rhodopsin and biologically altered cells that do rudimentary operations
within the paradigm of traditional computation.
-
Computational Biology
This category includes any effort to solve
biological problems by the application of computational methods and tools
to model biology and solve detailed mathematical expressions of biological
behaviors. Examples include methods of calculating from first principles
such as Ab Initio, Monte Carlo or other simulation programs applied to
looking at protein-protein interactions, protein folding, drug binding site
elucidation, etc.
-
Bioinformatics
This category includes the application of data
management, data mining, data modeling and algorithmic techniques to
biological databases, such as genome databases and related sequencing
information. Examples include in silico models as a predictive method for
gene function and data mining for inferring and determining sequence
homology information.
-
Biological Computation
This category includes efforts to
determine how biology does information technology from the sub-cellular
level to the systems and population level. The three main categories that
are expanded below are in silico systems for fundamental understanding;
hybrid systems to reverse engineer the biology and systems biology at the
multi-cellular level and beyond.