Wednesday, March 18, 2009

The Computational Biologist Arrives.

Biology started off as a descriptive science. There was a lot of observation involved and description of observed data. That was the day of the Naturalist. He/She went afield with telescopes in hand or a magnifying lens perhaps and waited hours patiently for the glimpse of that rare species. With the classification of life into kingdom animalia and plantae, evolved the Botanist, concerned chiefly with plants, and the Zoologist, concerned chiefly with animals. Then one day Antonie van Leeuwenhoek directed his microscope to a spoiled food item and a whole new world of the microorganisms was discovered. As understanding about the variety of flora and fauna increased interest grew in the behavior of these organisms. This was still a descriptive science.

Observation of behavior lead to observation of other phenomena and Mendel was the first to explain the concept of heredity. I consider the time between the early 1900s to mid 1900s the golden age for biology. Not only was the structure of DNA, which by then was proven to be the information carrier, elucidated but also the mechanism of coding this information found out. This period saw the birth of the experimental biologist. No longer did the biologist have to roam about in the fields, he explained fundamental biological phenomena in the laboratory. However unlike the experimental physicist who first builds a hypothesis, and then designs experiments to verify it, the experimental biologist has no hypothesis. Instead the experimental biologist asks objective questions about a particular phenomena and then designs experiments which will give him the answers. He uses these answers to then explain the phenomena. However the tricks and tools of his trade were not refined and were unwieldy.

Then came the revolution called molecular biology. This science is predominantly tool and technology based. The molecular biologist again works differently than the experimental biologist. The molecular biologist generates data which he then analyses to come to his answers. As the technology of molecular biology matured, the molecular biologist soon started drowning in the flood of data, in stepped the Bioinformatician to the rescue.

The main job of the bioinformatician was to arrange the data in meaningful structures which could be made sense of by other biologists. Bioinformaticians soon became indispensable, especially with the advent of the Internet and its associated services. With the Human Genome Project and the subsequent easy availability of data in public databases, understanding of biology progressed by leaps and bounds.

However even with the advances in molecular biology techniques many questions remain unanswered. A new school of thought has emerged which argues that further understanding of biology will only take place if the whole organism is studied as a single system rather than understanding different phenomena independently. The Systems Biologist tries to take a broader view of the problem at hand. Unfortunately with the level of detail known today, multiple phenomena can be analysed only by the modern day computers. The number of variables involved is so high that the human brain cant analyse it.

This is when the computational biologist arrives. He is an expert in handling computers. The Computational biologist not only needs to understand the softwares being run on his machine but also the hardware that his machine is made up of. The computational biologist, infact, uses his machine as an extension of his own mental capacity to solve a problem. He is an expert in programming and biology at the same time. Computational Biology is the future of Biology which will solve problems of a global nature, where the whole organism is involved. This revolution might not come about overnight. The level of complexity known today is so much that even superfast computers take a lot of time to simulate a protein-protein interaction for a small time interval, as much as 1 day of computation to calculate a time interval of 1 pico sec. None the less trying to solve these problems experimentally is, in some cases impossible, and in other cases, not financially feasible, especially now that funding is so hard to come by. The Computational Biologist has definitely arrived on the scene and with advancement in computer hardware and software, tough biological problems will be tackled only by the Computational Biologist.

The inspiration for this article comes from:
http://www.geocities.com/letapk/physics3.html

2 comments:

Vidhi Patel said...

hey nice one... sort of takes us on a ride thru evolution of biological sciences.
but computational biology still needs to be backed up by wet-lab experimentation, to an extent. i dont think that the programs will be able to interpret watever new information will be fed into it. some sort of basic info needs to be there, which can come only through experimentation.
but handling all that information systematically is definitely out of hands of humans alone.

Siddharth a.k.a. Plasmabhai said...

yes absolutely!!! well said!!! Afterall every program needs some input to analyse and generate the output. Infact for computational biology a pre-requisite in most cases is biological data generated in wet lab experiments.... However this is not absolutely compulsory, for example a program can be written which simulates evolution and to do that it only needs certain rules which govern evolution. Yes it does need that first of the Lowest Ultimate Common Ancestor (LUCA), or that perfect chemical solution, as an input, and this need not be experimental.