Computational biology: explore the magic of life by the power of computing

Life science is such a field that takes high respect of experiment data. With the fast progress in high-throughput genomics sequencers, huge amounts of sequencing data are accumulating, which brought with big opportunities for treatment of many tough diseases while how to dig valuable knowledge from these data will need help of computational biology.

As we mentioned above, the first step to explore the life system is to understand the building blocks, such as the genome. Gene, as a sequence of four words “ATCG”, contains the rules to guide the development of proteins, which construct the life system and modulate the running of the system. The genome, as a sequence of genes, is the container of these rules.  In order to obtain the accurate genome sequence, biologists designed a set of machines to extract DNA from tissues and then truncate them into small parts and record the code in each part. Just like Jigsaw puzzle, the recorded parts need to be connected to each other to present the whole picture of the genome. For a normal plant, such as wheat, we need to analyze about 5 Giga Byte data of parts to extract a genome sequence with around 200 Mega Byte size. It is far beyond humans’ ability and the only solution is to depend on computational software and algorithms to store the data and perform the analysis automatically.

sequencing to identify life secret (ref: http://en.wikipedia.org/wiki/RNA_sequencing)

Based on the accurate genome, we can further study the evolution of living system. For example, by comparing the genome sequence of people living in one area with others across the world, we can find some specific features which might do help to study the specific hereditary disease in that area. This kind of research will benefit from some existing computer science research such as text mining. Also some state-of-art database techniques such as Hadoop has also been employed for storing huge data and assist parallel data access in an efficient way[1]. 

As a dynamic system, the living organism will always maintain a complex metabolism network, which is consisted of a set of chemical reactions and its related modulator as enzymes. Computational biology can also help to simulate such a system with a network model. To put it simply, each node in the network will represent the reactant and product and each edge will represent a reaction. The reaction process can be modeled as differential equations and the concentration of reactants and products will be the variables. By setting the suitable boundary conditions to the network model, we can simulate the whole metabolism system. More interesting research is to simulate the result after manual intervention. Without a long time waiting for the wet-lab experiments, we can predict the organism’s behavior after some modulation on the environment conditions or the enzyme types, which takes great benefit of the computational model and algorithms for the simulation.

What I listed here are only small parts of examples that computing can be used in biology research. I believe that this is a promising research area which is just at its beginning.

Reference

[1] Charles Schmitt, UNC Big Data Analytics Stories: Genomic Sequencing, http://www.intel.com/content/www/us/en/big-data/renci-peer-story.html