William Bialek and Thomas Gregor
(for lectures at the 2007 Boulder Summer School on Condensed Matter Physics)
One of the most beautiful phenomena in nature is the emergence of a fully formed, highly structured organism from a single undifferentiated cell, the fertilized egg. Over the past decades, biologists have shown that in many cases the “blueprint” for the body is laid out with surprising speed and is readable as variations in the concentration of particular molecules (the “expression levels” of particular genes). In the fruit fly, we know the identity of essentially all the relevant molecules. As we try to understand how these molecules interact to form the patterns that we recognize as characteristic of the mature organism, we face a number of physics problems:
How can spatial patterns in the concentration of these molecules scale with the size of the egg, so that organisms of different sizes have similar proportions?
What insures that the spatial patterns are reproducible from one embryo to the next?
Since the concentrations of all the relevant molecules are small, does the random behavior of individual molecules set a limit to the precision with which patterns can be constructed?
Although the phenomena of life are beautiful, one might worry that these systems are just too complicated and messy to yield to the physicists’ desire for explanation in terms of powerful general principles. For the past several years, a small group of us have been struggling with these problems in the context of the fruit fly embryo. To our delight, we have been able to banish much of the messiness, and to reveal some remarkably precise and reproducible phenomena. In particular, the first crucial step in the construction of the blueprint really does involve the detection of concentration differences so small that they are close to the physical limits set by the random arrival of individual molecules at their targets. This problem may be so serious that the whole system for constructing the blueprint has to be tuned to maximize how much signal can be transmitted against the inevitable background of noise, and this idea of tuning or optimization can be turned into a precise theoretical principle from which we can actually predict some aspects of how the system works. Parallel questions of noise, reproducibility and information transmission arise in many different biological systems.
Our goal in these lectures is of course to give a sense of what we have learned about one particular problem, but also we’d like to give some flavor of how one confronts a biological phenomenon and recognizes the underlying physics problems. Hopefully it will be clear that one of the really fun parts of this effort is the interplay between theory and experiment.
This is very much work in progress.
Diffusion and scaling during early embryonic pattern formation. T Gregor, W Bialek, RR de Ruyter van Steveninck, DW Tank & EF Wieschaus, Proc Nat’l Acad Sci (USA) 102, 18403-18407 (2005).
Stability and nuclear dynamics of the Bicoid morphogen gradient. T Gregor, EF Wieschaus, AP McGregor, W Bialek & DW Tank, Cell 130, 141-152 (2007).
Probing the limits to positional information. T Gregor, DW Tank, EF Wieschaus & W Bialek, Cell 130, 153-164 (2007).
For background on the issues of noise and physical limits to the detection of small concentration differences, see these earlier theoretical papers:
Physical limits to biochemical signaling. W Bialek & S Setayeshgar, Proc Nat’l Acad Sci (USA) 102, 10040-10045 (2005); physics/0301001.
Cooperativity, sensitivity and noise in biochemical signaling. W Bialek & S Setayeshgar, q-bio.MN/0601001 (2006).
In a sense this theoretical work provided motivation for experiments which could sharpen the issues Now we can close the loop, and go back to some of these theoretical questions …
The role of input noise in transcriptional regulation. G Tkacik, T Gregor & W Bialek, q-bio.MN/0701002 (2007).
Information flow and optimization in transcriptional regulation. G Tkacik, CG Callan Jr & W Bialek, arXiv:0705.0313 [q-bio.MN] (2007).