The Cellular CPU
05/03/12 15:52 Filed in: GenomeWeb Daily Scan
Submitted by S. Pelech - Kinexus on Mon, 03/05/2012 - 15:52.
It is amusing to see a business magazine promoting to the general public hare-brain ideas from computer software programmers about how to genetically engineer cells by mucking with their DNA to create molecular logic circuits. While sophisticated physiological regulatory systems have indeed evolved within living cells, these are incredibly complex and operate at multiple levels from pre-transcriptional to post-translational. Our existing regulatory systems already have the capacity to recognize pathological conditions, and for the most part, they work exceptionally well. The best strategy to improve the delivery of human health care is to understand and either augment or repair these regulatory controls.
The concept of the focused development of man-made DNA software code that adjusts the levels of molecules in a "computational soup" to recognize and fight disease is extremely far-fetched, impractical and reflects incredible naivety about basic cellular and molecular biology. It may be possible to design DNA-based computers that solve the most simple of problems within test tubes or in in silico simulations, but not for the applications contemplated in this Economist article. For example, up to tens of thousands of genes in most genomes have co-evolved to work together, and there appears to be negative selection pressures to avoid changing DNA sequences from mutation or the introductions of new genes from viral insertions. The incorporation of new genes for DNA-based computers could actually have negative repercussions on the operations of normal genes resulting in very unhealthy cells.
Link to the original blog post.
It is amusing to see a business magazine promoting to the general public hare-brain ideas from computer software programmers about how to genetically engineer cells by mucking with their DNA to create molecular logic circuits. While sophisticated physiological regulatory systems have indeed evolved within living cells, these are incredibly complex and operate at multiple levels from pre-transcriptional to post-translational. Our existing regulatory systems already have the capacity to recognize pathological conditions, and for the most part, they work exceptionally well. The best strategy to improve the delivery of human health care is to understand and either augment or repair these regulatory controls.
The concept of the focused development of man-made DNA software code that adjusts the levels of molecules in a "computational soup" to recognize and fight disease is extremely far-fetched, impractical and reflects incredible naivety about basic cellular and molecular biology. It may be possible to design DNA-based computers that solve the most simple of problems within test tubes or in in silico simulations, but not for the applications contemplated in this Economist article. For example, up to tens of thousands of genes in most genomes have co-evolved to work together, and there appears to be negative selection pressures to avoid changing DNA sequences from mutation or the introductions of new genes from viral insertions. The incorporation of new genes for DNA-based computers could actually have negative repercussions on the operations of normal genes resulting in very unhealthy cells.
Link to the original blog post.