An article, “Why Do Dachshunds Have Short Legs? Science Has an Answer,” in R&D, July 17, 2009, reminds me of an observation. In the language of computer programming, a retrogene is a patch on a pre-existing piece of software. Could it be that the entire embryogenic/genomic network that dictates a creature’s morphology is nothing more than a set of onion-skin layers or patches on conserved coded-machinery that has worked before? If so, it’s going to take a lot of industrial-strength gene-insertion genomics to unravel it, since it has no real logic that would help us make sense of it and guide us to a proper reading frame.  (This is more evidence that there’s no “intelligence” in the “Intelligent Design” of Darwinian evolution.)

We need to devote significant capital resources to this task of Synthetic Biology as soon as we can, so we can decipher the mouse genome as fast as we can. Remember that there are only ~25K genes. On the other hand, there are a lot of junk-DNA/RNA targets to figure out as well, so it’s not going to be easy.

Will someone like Craig Venter come to our rescue with a “just so” methodology? Let’s start by snipping out big chunks of DNA systematically to see if the results are (1) lethal (no development); (2) invisible (normal mouse); or (3) visibly pathological (apparent congenital defect). Then, for the third case, divide and conquer to figure out what each chromosomal sub-segment does to give rise to the observed defect.

This process may not converge in our lifetimes if we don’t get started right away with an approach that automates the process with a robotic arm(s) and puts the mouse embryos in a 10K-well glass/plastic microarray with proper growth media that can be visualized with computer image-processing software over several days to determine which zygotes are undergoing development and which can be ignored. Then, we need to insert the developing blastocysts into a receptive surrogate female mouse to see what happens next during the course of a normalpregnancy.

This project will take lots of resources, but I don’t see how else to do it in a mammal that will yield the homologous genetic information that will be relevant to us humans in a reasonable amount of time.

Any ideas?

– L. Stephen Coles, M.D., Ph.D., Co-Founder
Los Angeles Gerontology Research Group
http://www.grg.org