What the scientists told Allen was that they really wanted a virtual map for the genes that turn on and off in the brain, akin to the blueprint of the entire human genome that, at that time, was nearing completion. So in 2003, the same year that the Human Genome Project (HGP) was finished, Allen pitched in $41 million and launched the Allen Brain Atlas, an ambitious — and altruistic — indexing of the entire genome of the mouse brain that would be available, free of charge, to researchers on the Web. Why mice? It's impossible to get the live samples of human brain neurons needed to map the human brain genome in the same way, but people and mice share 90% of brain genes, making the mouse a pretty good stand-in.
The project, which was completed Tuesday, details the activity, or expression pattern, of genes. It includes vivid three-dimensional images and descriptions of 20,000 genes and their activity, and has already become a go-to source for researchers studying everything from multiple sclerosis to brain tumors. Researchers can log on and view, in helpful color-coded patterns, where certain genes are turned on in the brain, and can manipulate the image to get just the perspective and cross section they need.
For Dr. Greg Foltz, a neurosurgeon at the Seattle Neuroscience Institute at Swedish Medical Center who studies incurable brain tumors, the mouse atlas functions as a springboard for better understanding how these difficult tumors develop and grow. "We need clues," he says. "When a patient comes in and has a tumor removed, we take that tumor and complete a genomic study, but all we have is a database of genes. The best analogy I can come up with is that this genomic data is like having just the names in a phone book; it's only a list. We want to know what those genes do. So we can now go to the mouse atlas, which gives us the gene [activity] pattern in a normal mouse brain. Is this gene expressed in normally? Is it expressed at a higher or lower level in a tumor? We use the atlas every day, to figure out which genes are important to the biology of the tumor, and which are bystanders."
Foltz is also part of the Allen Brain Atlas' next project, which involves mapping the human cortex. The cortex is where sensory information from the eyes, ears, nose and touch neurons is processed. He and his group are providing samples of human brain tissue, including archived material from patients who have suffered from diseases such as epilepsy, to give researchers a better understanding of which genes are involved in disease states.
The Atlas is just the start of what neuroscientists see as the future of brain science. Already, the mouse atlas has revealed something new about brain neurons — researchers had always thought that brain neurons were pretty much the same, and were distinguished mainly by how they were hooked up to each other. It turns out, however, that brain cells, like other cells in the body, show a diverse array of gene expression patterns, meaning that different cells use different genes at different times to perform their complex functions. "Under the microscope, neurons look the same, but in terms of gene expression, it looks like there are differences," says Foltz. "This has opened up whole new fields for cognitive neuroscientists to understand what makes the [neurons] that go to the eye different from the ones that go to the memory center or the ear. The atlas allows you to make inferences, and inspires you to think of all kinds of things."