Communications of the ACM
Communications of the ACM
Computers and Biology
Can we replace costly and possibly unethical animal experiments by computer runs? Can we have computer versions of ourselves that can be used for medical experiments, reducing the danger of risky trials? I believe will have some positive answers to these questions in the centuries ahead.
It is safe to infer the major role that computer science will play in molecular biology for future generations. Both deal with discrete structures (for example, four nucleotides, 20 amino acids, triplets representing codons) and both also deal with a flow of control that regulates the generation of proteins (for example, operons), much like semaphores dictate program execution. The growing interaction between the two disciplines is reflected in the recently coined term "bio-informatics."
During the last two centuries mathematics played a key role in physics because it provided the foundation for establishing new theories. Computer science will play a key role in biology in the centuries ahead because those mathematical theories are mostly unsuitable. For example, the human genome project would be unfeasible without computers, and that task is just a minuscule fraction of what remains to be accomplished to comprehend the behavior of a cell.
Computers will be essential to simulate cells both at the micro and macro level. In the micro stage, one studies structure at the atomic level and must learn how proteins fold and interact. Proteins are the building blocks of every living organism and they are generated from DNA in a machine-like manner. The unique 3D geometrical shape of a given protein is a key element in inferring that protein's function.
The study of protein folding and protein interactions will require huge computational resources. I already anticipate a revival of massively parallel machines. Indeed, the IBM team that designed the chess-winning program is now developing computers with millions of processors to study protein folding.
At the macro level, dealing with more abstract structures representing large biochemical compounds, an understanding of cell regulation and metabolism becomes paramount. Micro-arrays or DNA chips are already used for that purpose. From a computer science perspective, huge amounts of data will be generated and require interpretation. Data mining and machine learning will play key roles in those endeavors.
Once researchers succeed in simulating a cell or clusters of cells, then they can supersede in-vitro and in-vivo experiments by in-silico experiments. The next step is to provide answers to the two questions asked at the outset of this essay. And let's add a more difficult one: How will humans evolve?
Another aspect of biology that may shape future computers is the potential for miniaturization and generation of exponential numbers of biological processors. The era of DNA computing has barely started and it is already changing the well-established notions of complexity. It is not inconceivable that biological computers will open new vistas in computation.
Interdisciplinary research demands efforts from the disciplines it congregates. Educating our students in this fascinating new area will bring us closer to understanding ourselves.
When men speak of the future, the gods laugh.
Chinese proverb
The Digital Library is published by the Association for Computing Machinery. Copyright © 2001 ACM, Inc.
No entries found