Research Roundup

Development is easy

An actual lineage (left) is much simpler than a randomly generated lineage (right). AZEVEDO/MACMILLAN

Metazoan cell lineages can be collapsed to a set of rules that is surprisingly simple, according to Ricardo Azevedo (University of Houston, Houston, TX), Armand Leroi (Imperial College, Ascot, UK), and colleagues. Decoding the biochemical basis of the rules should provide a complete recipe book for development.

Azevedo brought three influences to the study: his background in evolutionary biology; his adopted field of worm biology; and computer science from collaborators. When he joined Leroi's worm lab he "was immediately struck by the lineage data," he says. "But there were more data to be extracted."

As in phylogenetic trees, there were repeating patterns. A handful had been noted by others, but no systematic study had been undertaken. This is where a simple computer algorithm helped out.

"For any particular cell division we can find another that forms the same pattern," Azevedo explains. "Then we collapse [those into one rule] until we can't do it any more because there is no more redundancy. That gives us the minimal number of states that is required."

In silico evolution of the simplified rule sets did not yield much further simplification, as long as the final distribution of cell types was constrained. Much simpler rule sets could be invented but only by going via intermediate states that had very different cell type distributions.

Azevedo thinks that simplicity arises as evolution strives to minimize the time and genetic information necessary for development. The tendency of evolution to modify what already exists, rather than invent new systems for each new function, may also favor simplicity.

In silico genetic circuits that generate lineages are allowing Azevedo to study the rules behind lineage formation. Eventually he hopes to understand what biochemical combinations of regulators form the basis for each of his rules, but that may have to wait for next-generation expression chips that can analyze individual cells.

Reference:

Azevedo, R.B.R., et al. 2005. Nature. 433:152-156. [Abstract]

William A. Wells

wellsw{at}rockefeller.edu
CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This Article Alert me when this article is cited Services Email this article Similar articles in this journal Alert me to new content in the JCB Download to citation manager Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Search for Related Content Social Bookmarking                            What's this?