Just when you thought that The Dancing Wu Li Masters and The Tao of Physics had finally been left in the 1970s, along comes a surfer living on the Hawaiian island of Maui who claims to have a "simple theory of everything" that shows that the universe is an "exceptionally beautiful shape." Garrett Lisi has a physics PhD but no university affiliation, and lists his three most important things as physics, love and surfing (not in that order).
But Lisi is no semi-mystic drawing charming but unedifying analogies. He is being taken seriously by the theoretical physics community, and has been invited to the high-powered Perimeter Institute in Waterloo, Canada, where physicist Lee Smolin has called his work "fabulous."
Lisi's work is almost comprehensible—at least by the standards of modern fundamental physics. In comparison to string theory, the main contender for a theory of everything, Lisi uses only "baby mathematics." But this is not to say that it's easy.
A theory of everything must unify the theory of general relativity, which describes gravity and the structure of spacetime on large scales, with quantum theory, which describes how particles behave at the subatomic scale. To put it another way, gravity must be mixed into the "standard model" of particle physics, which explains the interactions between all known fundamental particles—quarks, electrons, photons and so forth.
Physicists typically attempt unification by using symmetry. In a crude analogy, two particles that look the same except for the fact that they spin in opposite directions can be "unified" into a single particle by appreciating that they can be interconverted by reflection in a mirror—a symmetry operation.
The idea is that the proliferation of particles and forces in today's universe is the result of a series of "symmetry-breaking" steps, just as stretching a square into a rectangle reduces its symmetry, converting four identical sides into two distinct pairs. This is already known to be true of some forces and particles.
Lisi claims that the primordial symmetry is a pattern called E8, known to mathematicians for over a century but only recently fully understood; it is like a multidimensional polyhedron with 248 "corners." He has shown that all of the known particles, plus descriptions of gravity, can be mapped on to the corners of E8. So part of it looks like the standard model, while another part looks like gravity and spacetime. Twenty of the "corners" remain empty: the model thus predicts the existence of hypothetical particles. It is like the way 19th-century chemists found a pattern—the periodic table—that brought coherence and order to the chemical elements; the gaps in it predicted elements that were later found.
Is E8 really the answer to everything? Physicists are reserving judgement. Lisi's paper, which is not yet peer-reviewed or published, is just a sketch—not a theory, and barely even a model. Mathematical physicist Peter Woit is unsure, saying that playing with symmetry just defers the question of what breaks it to make the world we know. But the trick worked before in the 1950s, when Murray Gell-Mann predicted a new particle by mapping a group of known ones on to a symmetry group. Either way, Lisi's suggestion invigorates a field that, wandering in the thickets of string theory, sorely needs it.
From skin cell to stem cell
Stem-cell researchers in Shoukhrat Mitalipov's team at the Oregon Health and Science University might be forgiven a little chagrin. No sooner had they reported a major breakthrough than they were trumped by two reports apparently offering an even more attractive way of making human stem cells.
Stem cells are the all-purpose cells present in the very early stages of embryo growth that can develop into just about any type of specialised tissue cells. The "traditional" strategy for giving them the DNA of the eventual recipient involves stripping the genetic material from an unfertilised egg and replacing it with donor DNA. Then the egg is prompted to grow into a blastocyst, the initial stage of an embryo, from which stem cells can be extracted. This is called somatic cell nuclear transfer (SCNT), and is the method used in animal cloning. It works for sheep, dogs and mice, but there had previously been no success for humans or other primates.
On 14th November, Mitalipov and colleagues reported they had made stem cells by SCNT from rhesus macaques. But a week later, teams based at the universities of Kyoto and Wisconsin-Madison independently reported the creation of human stem cells from skin cells. The cells were treated with proteins that reprogrammed them, switching the gene circuits from a "skin cell" to a "stem cell" setting. This reversal of normal developmental pathways is extraordinary.
The two teams used different cocktails of proteins, showing that there is scope for optimising the mix. Best of all, the method avoids the creation and destruction of embryos that has dogged the ethics of stem-cell research. Mitalipov insists that starting with eggs is still best, and he has started collaborating with a team in Newcastle licensed to work with human embryos. After years of effort, all options suddenly seem open.
