Dennis Wall: From moss to autism

By Kelly Rae Chi

© 2009 Leah Fasten

In 1996 a few hundred plant scientists gathered in Baton Rouge, La., for an annual phylogenetics meeting. Biology undergraduate Dennis Wall rushed into a lecture hall to meet Brent Mishler, a University of California, Berkeley, integrative biologist, who was considering taking on Wall as a doctoral student. But when Wall, late and disheveled, tried to climb over a row of folding chairs to get a closer seat, he tripped, and the chairs crumbled beneath him. When he got up, applause filled the auditorium.

Researchers including Mishler, who became Wall's PhD advisor, say Wall would continue to surprise them—academically, that is—for years to come.

As a PhD student, Wall tracked the radiation of an obscure genus of moss from its origin in Malaysia to its outermost ranges in French Polynesia. He wanted to see how quickly branches...

Wall applied the epidemiological model to genetic data and showed that the moss diversified at a blistering evolutionary pace.1

As a postdoc in biologist Marcus Feldman's Stanford University lab, Wall analyzed more than 3,000 proteins in four yeast species, and showed that both the expression and "dispensability" of a protein can affect the rate of its evolution, something that researchers suspected but few had convincingly shown.2

In 2003, Wall became the director of Harvard Medical School's Computational Biology Initiative. There he created an online content management system that integrated tools used by many HMS researchers to analyze DNA and amino acid sequences. In April 2006, Wall took a faculty position at the newly formed Center for Biomedical Informatics at HMS. He faced a number of challenges, says Isaac Kohane, associate professor of pediatrics who recruited Wall and formed the center. Wall wasn't trained in medicine. He had to figure out which questions he could answer using phylogenetic techniques, and he had to harvest the right data for these meta-analyses. "I have to admit that I was silently skeptical," Kohane says. "Boy, was I wrong."

Drawn to autism in part for its biological complexity, Wall and his group began mining the literature and integrating disease determinants—genes, transcription factors, and proteins—that could help generate a network of autism gene candidates.

Related Articles

3 Wall says he hopes that the system will eventually "flatten the playing field" so every researcher can use sophisticated informatics.

Wall is now working on several projects that stem from Autworks, including one that uses calculations underlying game theory to decode patterns of gene expression in autism patients. This type of analysis compares microarray data from autistic patients and might uncover disease gene candidates missed in older comparisons that couldn't capture potential gene candidates amid the variation in gene expression across the spectrum of autism sufferers. The theory, he says, has the benefit of discerning the contributions of individual genes involved in autism while considering contributions from other genes in the network. "It's sort of an economic analysis of the autism genescape," he says, where the "market" is the highly variable transcriptional readouts from people with autism. "It's a first step, but potentially an important one if we have any hope of using mRNA expression data to reveal diagnostic signatures indicative of autism."

Title: Faculty, Center for Biomedical Informatics, Harvard Medical School, Boston
Age: 34
Representative publications:

1. D.P. Wall, "Origin and rapid diversification of a tropical moss," Evolution, 59:1413–24, 2005. (Cited in 9 papers)
2. D.P. Wall et al., "Functional genomic analysis of the rates of protein evolution," Proc Natl Acad Sci, 102:5483–88, 2005. (Cited in 76 papers)
3. D.P. Wall et al., "Comparative analysis of neurological disorders focuses genome-wide search for autism genes," Genomics, 93:120–29, 2009. (Cited in 0 papers)

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!
Already a member?