Experimental Evolution

	The fitness of E. coli competing for lactose can be predicted solely from a knowledge of biochemistry using Metabolic Control Theory and Biochemical Systems Theory. The surface is the fitness predicted from a knowledge of enzyme kinetics at each step in the lactose pathway. The colored dots are data. Dean. 1989. Genetics 123:441-454. Our work shows that the phenomena of epistasis (additive, synergistic, and antagonistic) dominance, modifiers of dominance, conditional neutrality, and GxE interactions are the inevitable consequences of the simplest of metabolic architectures. Dykhuizen & Dean. 1990. TREE 5:257-262. Dean. 1995. Genetics 139:19-33.

	Fitness across an environmental gradient can be predicted solely from a knowledge of biochemistry. Two E. coli strains carrying different lac operons, one favored on methylgalactoside (TD10) and one favored on lactulose (TD2), are predicted to form a balanced polymorphism within a narrow zone on the gradient. Dots are data. Lunzer, Natarajan, Dykhuizen & Dean. 2002. Genetics 162:485-499.
	.
	Further adaptation was expected to destabilize the balanced polymorphism. Mutations affecting fitness on the pure sugars were expected to shift the narrow zone of coexistence away from the resource supply.
	.
	Mutations dramatically affect fitness. Amazingly, evolved TD10 (2261) is fitter on lactulose and evolved TD2 (2262) is fitter on methylgalactoside and the zone broadens. Dykhuizen & Dean. 2004. Genetics 167:2015-2026.
	.
	We seek to identify the mutations causing changes in resource specialization and to understand the biochemical basis of the evolution of flipped balanced polymorphisms. Zhong, Kodursky, Dykhuizen & Dean. 2004. PNAS 101:11719-11724.