Stanford University School of Medicine, California, USA

A systems biologist encourages modelling by the millions.

In a typical modelling study, we write down equations, solve them, and see whether they account for known data. If they do, we claim to understand some bit of biology. One huge caveat is that many other models might have matched the data just as well.

Researchers from Peking University in Beijing and the University of California, San Francisco, have devised a satisfying way of dealing with this problem (W. Ma et al. Mol. Syst. Biol. 2, 70; 2006).

Their starting point was epithelial patterning in the fruitfly Drosophila. During embryogenesis, a system known as the 'segment polarity network' generates repeating stripes of gene expression. The stripes are initially fuzzy and later become sharp. Ma et al. set out to see what simple gene circuits were best suited to this sharpening process.

They formulated differential-equation models for about 14 million ways of connecting two or three segmentation genes, then randomly chose 100 sets of parameters that defined the strength of the interactions for each gene. They then carried out computations for each combination to determine which of them converted fuzzy stripes into sharp ones.

Many topologies worked for at least one parameter set. But only a fraction worked for more than one or two. Interestingly, the most robust topologies were all variations on the same design — each had three sub-circuits, one 'stripe generator' motif and two bistable 'response sharpeners'. These findings give hope that complex networks may be decomposed into modular sub-circuits with understandable functions.

Comprehensively examining millions of models is a lot of work, but is not impossible. And, as Ma et al. show, it can yield important insight that could not have been derived from studies of one or two.

Posted by Brendan Maher at 07:54 PM | Permalink | Comments (1) | TrackBacks (0)

Korea University, Seoul, Korea

A spectroscopist tells how the tools of his trade are revealing quantum effects in biological molecules.

In introductory quantum mechanics, one learns that particles can behave like waves, with each particle having a wavelength inversely proportional to its momentum. I am fascinated by recent work that examines how, at the molecular level, life takes advantage of these wave-like properties.

The effects aren't visible when moving biomacromolecules are viewed whole, because their large mass means they have a negligibly small 'de Broglie' wavelength. However, atoms and electrons within a molecule — for example, in active sites, where reactions such as catalysis and light-absorption take place — may interact in a wave-like way.

Researchers are thus investigating what role the wavefunctions of these molecular constituents have during biochemical reactions. And if they do interact, over what distance and for how long does the wavefunctions' phase relationship, or quantum coherence, persist?

Spectroscopists have found evidence for coherences in a few biological systems, thanks to a technique known as multidimensional spectroscopy (A. Nagy et al. Curr. Opin. Struct. Biol. 16, 654–663; 2006). This involves tracking changes in a molecule's configuration over very short timescales with laser pulses that last femtoseconds (10-15 s).

Further results reported this year suggest that the energy transfer in a photosynthetic system is wave-like (G. S. Engel et al. Nature 446, 782–786; 2007). For this process, the quantum coherence of the light-excited charges may help the charges search out an efficient pathway through the molecule, by means of a mechanism analogous to a quantum computation.

This observation provokes a question that I look forward to seeing answered. Might biological systems have evolved to use matter's wave-like properties to optimize their efficiency?

Posted by Brendan Maher at 10:27 PM | Permalink | Comments (0) | TrackBacks (0)

University of Manchester, UK

Imploding atoms have softened this experimentalist's teasing views on theoretical physics.

As an experimentalist, I instinctively dislike theory papers. Too many of them seem to be written for the sole purpose of showing off an integral larger than a competitor's, or to present multiple theories just in case one idea proves right and so is hailed as visionary. I feel even less warmly towards theories that are nigh on impossible to check, such as the supposed precursor to a theory of everything, string theory.
But speaking seriously, even the most obscure predictions can turn out to be spectacularly relevant.
In our lab we have been studying graphene, a material that comprises a single layer of carbon atoms arranged similarly to chicken wire. Because electrons in this material mimic ultra-relativistic particles, it should be possible to observe in their behaviour century-long-predicted phenomena such as the Klein paradox (which concerns how highly energetic electrons tunnel through supposedly impenetrable barriers) and zitterbewegung (jittery movements of relativistic wave-packets).

Several recent theory papers on the physics preprint server arXiv predict another coup for graphene (see A. V. Shytov et al. arXiv:0708.0837; 2007).

According to relativistic quantum theory, atoms containing more than 170 protons cannot exist, because electrons around nuclei with such a large charge would fall into the centre. Nuclear physicists have not come close to creating atoms heavy enough to test this prediction. But the recent theory papers suggest that it should be relatively easy to observe the effect in graphene. This is because electrons in this material interact much more strongly than they do in atoms, so should fall down on charged impurities (standing in for nuclei) rather routinely.
This makes me wonder: could we design condensed-matter systems to test the supposedly non-testable predictions of string theory too?

Posted by Jenny Hogan at 12:06 PM | Permalink | Comments (2) | TrackBacks (0)

Ecole Normale Supérieure de Lyon, France

A geochemist goes à la recherche des climats perdus.

As a young postdoc at the California Institute of Technology (Caltech) in Pasadena I remember glancing through the 1952 logbook of a gas mass spectrometer while the machine readied my samples. In the book, Sam Epstein, one of the founders of modern geochemistry, had scribbled numbers representing the first attempt to determine past temperatures from oxygen-isotope abundances in fossils.

Since Epstein's measurements, the abundance of oxygen-18 in the carbonate skeletons of fossil sea creatures has become a broadly used indicator of past ocean temperatures. Such data are key to understanding modern climate change. But the usefulness of 18O in 'palaeothermometry' is limited by problems including variations in oxygen-isotope levels in sea water and in the way different organisms take up the isotopes.

Recently, a group at Caltech proposed a measurement that may work better. As before, the carbonates are broken down into carbon dioxide for analysis. Instead of looking only for molecules containing 18O, the Caltech team measures the abundance of molecules that contain both 18O and the uncommon carbon isotope, carbon-13. The excess of this species over what would be expected through random combination of carbon and oxygen atoms indicates the temperature at which the carbonate formed.

Early tests of this 'clumped' thermometer on corals and fish ear bones were promising (P. Ghosh et al. Geochim. Cosmochim. Acta 70, 1439–1456; 2006; and Geochim. Cosmochim. Acta 71, 2736–2744; 2007). Since then, the method has provided a new record of ocean temperature during the Palaeozoic era, which began 543 million years ago (R. E. Came et al. Nature 449, 198–201; 2007).

I believe that clumped isotope thermometry is going to be a valuable new tool for palaeoenvironmental studies.

Posted by Jenny Hogan at 07:11 PM | Permalink | Comments (6) | TrackBacks (0)

University of Chicago, Illinois, USA

An evolutionary geneticist is surprised by genes of unknown origin.

I once thought that, like us, every gene must have a mother. But recent work has identified some genes that seem to have no genetic ancestry. These 'motherless' genes pose a new challenge to understanding the molecular mechanisms and evolutionary forces that shape our DNA. This isn't the first time we've had to revise our ideas about gene evolution.

About 40 years ago, geneticist Susumu Ohno proposed that new genes originate when an existing gene duplicates, then one of the copies evolves a new function. Working with Chuck Langley in the early 1990s, I had the luck to discover a gene in flies that added another strand to Ohno's story. The gene, named Jingwei, is a chimaera that formed through the combination of two existing genes.

Since then, researchers have identified many other 'new' genes assembled from unrelated genes and mobile DNA elements. Often the sequences' origins can be identified. When they can't, researchers have simply assumed that subsequent evolution has masked the relationship of the gene to its ancestral sequences.

But this is unlikely to be the case for hydra, a gene found recently in Drosophila melanogaster and closely related species (S.-T. Chen et al. PLoS Genet. 3, e107; 2007). No homologous sequences are found in a species that diverged from those carrying hydra only 13 million years ago — too recently for mutations to have obscured any related sequences. This implies that hydra arose de novo.

Another group has found a further 16 de novo genes in flies, which they propose evolved from non-coding DNA (D. J. Begun et al. Genetics 176, 1131–1137; 2007 and M. T. Levine et al. Proc. Natl Acad. Sci. USA 103, 9935–9939; 2006). These genes beg further study: what initiated their formation?

Editor's Note, the entry previously misspelled the name of the author's institution. Nature regrets the error.

Posted by Jenny Hogan at 06:00 PM | Permalink | Comments (10) | TrackBacks (0)