doi:10.1038/nindia.2010.145 Published online 22 October 2010
Alphabets and numbers are human inventions that do not exist in the real world and yet find use in construction of buildings and bridges. Language and its underlying grammar are born out of such needs.
Scientists asking similar questions on the grammar of the language that composes life have created a new computer lingo named 'Kera'. The language is an answer to simple posers — can we understand and represent life's grammar in the form of a computer language; if DNA represents the source code, how are different protein objects compiled in different situations?
The Centre for Systems and Synthetic Biology and the Centre for Bioinformatics at the University of Kerala have developed this open source language to construct biological systems from scratch. Kera (meaning coconut in Malayalam) is a new communication interface with biology and is based on computer aided design of organisms. The idea is to create a digital version of the 'Life Operating System', build predictive models and design novel applications.
The general purpose, object-oriented language is based on microsoft's C# framework. In the Kera environment, organisms are represented as 'classes', metabolic pathways as 'functions' and DNA-RNA-protein interactions as 'rules'. DNA is viewed as a source code that compiles into different RNA and protein molecules under various environmental conditions.
The challenge is to write a computer code that closely resembles biology. Each molecular interaction is stored in the form of rules in a library, called Samhita.
Project leader Umesh P. has abstracted cellular processes, modularizing the problem into stand alone components and working backwards to assemble modules into a virtual cell. Kera is designed to be an adaptive language. "It will learn over time as more data comes in", Umesh says. "The beauty of the Kera is that it will bring us closer to Nature."
The emerging area of computer aided design of organisms is a walk in the space of the unknown. The gene-inventory that makes bacteria like E.coli is reasonably well catalogued. However, it is unclear how these parts work together to generate life.
The big question remains: can we completely convert life into numbers? How do we find a numerical equivalent of consciousness? Does soul have a periodic table? Is DNA the Director of Nuclear Affairs or merely a database that stores life's instructions?
Recent scientific evidence points to the fact that biological processes are analog and non-linear. It would be interesting to see if Kera can help sort cellular decisions into stand-alone modules and re-assemble them into an organism. "Quite frankly, it is a long way off," says Achuthsankar Nair, the chief architect of the Kera project. "We do not know the rules composing organisms. But we will get there eventually".
The ultimate dream of its developers is to run executables of biological processes for process analysis and engineer biological systems towards novel and useful applications.