Published by biologyonline.com on March 29, 2009
Mathematics, statistics and computational methods are some
of the key tools used by bioinformatics scientists to unlock patterns in large
datasets. This is a bit like picking up a spanner and expecting to be able to
change a flat tyre. A spanner might help, but unless you know how the tyre fits
in with the car, or indeed what the spanner is capable of, then the act is
The realm of philosophy has been much maligned by science.
Philosophers get accused of navel gazing, while real scientists’ get on with
the hard work of empirical investigation. Empirical facts slowly reveal the
true nature of the phenomenon under investigation. On a day-to-day basis
science is a slow affair, but over the past 200 years much has been revealed by
this tried-and-true approach.
The Periodic Table of Elements was derived by understanding
the empirical properties of discovered elements and identifying common
patterns. The result: an abstract categorisation of the elements that has stood
the test of time. But still, there is no theoretical logic to guide us as to
how many elements there might be or, indeed, why electrons are organised the
way they are. Mathematical descriptions are not logical explanations. The
elements are fundamental components that constitute everything within the
universe. Unless we can understand why they are structured the way they are,
moving forward would seem destined to remain a slow affair.
The structure of the double helix was derived by
understanding a range of empirical evidence: the chemistry of the nitrogenous
bases, the way the nitrogenous bases bond, Chargraff’s data on the quantity of
the bases, and Franklin and Wilkins X-ray images of DNA. The result: an
abstract description that led to empirical evidence to confirm the B-form of
helix that has stood the test of time. A giant leap in biology ensued, and a
great deal has been uncovered about the many forms of life of earth. However,
biology appears to be at a standstill relying on bioinformatics that resembles
the process of trying to find a needle in the haystack.
A fundamental question that could release the floodgates
remains unanswered: What is the underlying theoretical logic for the features of
DNA? And, if there is an underlying logic, does it have any relationship with
the fundamental elements and can such a theory provide any value to
understanding the codons? With only 64 codons making 20 amino acids, a number
of scientists believe that there must be some underlying logic to these
The hierarchy of living things has been subject to many taxonomic
arguments. Every time new empirical evidence emerges, taxonomies must be
revised. The Periodic Table has been subjected to this on-off process. New
discoveries require alterations to the categorisation, then the abstract
categorisation remains stable for a while, but new evidence emerges and the categories
get altered again, and on we go. While
the Periodic Table is relatively stable today, discoveries of new elements may
yet give rise to a change in its shape.
What is missing from modern science today is the use of
theory. Trial-and-error science will not yield a paradigm shift. Mendeleyev
constructed a theoretical description of the relationship between the elements
based on a theory of periodicity. Watson and Crick built a theoretical model
for DNA that empirical evidence proved accurate. But, there appears to be no
one searching for a theoretical link between the elements (a single molecule)
and DNA (a macro-molecule); even though some scientists are searching for the
relationship between the nitrogenous bases and their triplicate cousins (the
Recently, Dr Glassop, from Deakin
has developed a theoretical model that provides an explanation for the
underling features of the Periodic Table and the underlying features of the
nitrogenous bases (see Glassop 2007). Dr Glassop’s model describes a
multi-faceted view of causality that gives rise to a multi-faceted ontological
schema. The model is referred to as the Structure-Organisation-Process (SOP)
model of change.
Most scientists would stand aghast at the idea that a causal
model might be useful to biological science. After all, causality stands
accused of having limited relevance to simple chemical systems. And, many have
doubts that causality has any relevance at the sub-atomic level. If causality
was to have relevance to complex biological life, then we might need to invoke
ideas of anthropomorphism; where all
forms of life act as we humans do with motivation and purpose. However, it is tautological to expect a
paradigm shift when the current paradigms cannot be challenged!
To suggest that a causal model can explain the structure of
a nucleotide base, the organisation of electrons in an atom and the processes of human behaviour
might render all things in the universe connected in some way. Although, if the
universe itself is considered a system, then, by definition, all things
contained within that single system must be connected! It is the job of science
to reveal these connections.
Dr Glassop’s (2007) view of the Periodic Table claims that
there are eight periods in a theoretical model (not seven), the second period
is left vacant (because of certain violations), the orbitals, shells, lobes and
spin are cumulative features, the numerical representation of the SOP model is
reflected in the quantum numbers (of particular interest is the rationale for
the lobes) and there is a limit to the number of elements we can expect to
discover (118). It is only with a theory of the underlying pattern for the
Table that so many ideas can emerge.
The SOP model confirms recent empirical evidence about the
nucleotide bases: cytosine (C) and adenine (A) give rise to the right twist of
the helix, while thymine (T) and guanine (G) give rise to the left twist (see Ha
et. al. 2005). Of special interest to bioinformatics is the representative
numbers provided by the SOP model (C=1, T=7, A=5, G=3). While accusations of
numerology have been levelled at Dr Glassop’s work, the 1-7-5-3 order is
reflected in the blocks of the Periodic Table according to the number of
orbitals (s=1, f=7, d=5, p=3).
The SOP model of change is not a predictive model and there
are no algorithms that can find immediacy in bioinformatics, but the theory is
novel and does deserve the scrutiny of scientists searching for the next valve
that releases the floodgate.
Deakin University, Melbourne,
Ha, S.C., Lowenhaupt, K., Rich, A., Kim, Y. and Kim, K. (2005). Crystal
Structure of a Juncture between B–DNA and Z–DNA Reveals Two Extruded Bases. Nature.
LI (2007) Rethinking Causality; pattern
as the science of change, Heidelberg Press, Melbourne.