Assistant professor of molecular biology Magnus Nordborg has received a $2.7 million grant from the National Science Foundation to study the natural genetic variation of a useless weed.
Along the way, he hopes to develop a powerful new technique for genetically analyzing plants and identifying important genes in vital crop species.
The unassuming plant, Arabidopsis thaliana, is so devoid of agricultural, pharmaceutical or even ornamental value that it doesn’t even have an English name. Yet as a “model organism” – a species with features that make it ideal for scientific study – it enjoys strong popularity among geneticists, who probe it for insights into basic plant biology.
The main goal of Nordborg’s three-year project, soon to begin in collaboration with colleagues at the University of Chicago, is to create a database that captures the genetic variability among 96 different strains of the weed from all over the world.
“The point is not to sequence the whole genome,” said Nordborg. “That’s already been done, though only in one strain. We’re going to re-sequence it – in 96 strains, getting the information from [just a few] points across the genome.”
Rather than exhaustively catalog all 130 million of the genome’s DNA “letters,” Nordborg intends to sequence small segments, just a few hundred DNA letters each, at fewer than 2,000 sites along the chromosome of each strain. This amounts to just 1 percent of the Arabidopsis genome. Yet he expects the database, once compiled, will contain enough information to become an important tool for so-called gene mapping. This is the process of locating genes by looking for statistical associations between interesting traits and landmarks on the chromosome called genetic markers.
This strategy is viable in Arabidopsis because it is self-fertilizing. Since each individual plant has been inbreeding with itself for countless generations – essentially shuffling identical copies of its chromosomes together – marker-trait associations are extensive in the genome.
“This should mean that with a relatively sparse map [of the genome], we should be able to relatively quickly zone in on the genes for important traits,” Nordborg said.
“If this actually turns out to be a powerful method for mapping genes, we’ll be able to basically map genes in silico [via computer]. The hope is that with such a database, you can just go in with your phenotype information” – observational data about specific traits, such as drought resistance or cold-hardiness – “and ask the computer where the genes are. You won’t need to do any crosses. You won’t need to genotype anything. You’ll just ask the computer.”
If it works for Arabidopsis, Nordborg said, it should also work for other self-fertilizing species, including commercially important ones like rice and barley. This would be a great boon to agriculture, increasing the speed and lowering the costs of identifying and selecting useful genes.
This project is the first-ever attempt to describe the genomic variation across an entire species, Nordborg said. The database will be made publicly available, he added, and should be a valuable resource for students of computational and evolutionary biology.
Nordborg earned his Ph.D. at Stanford University and subsequently did research at the University of Chicago and Lund University in his native Sweden. He came to work as a professor at USC less than a year ago. Nordborg is an affiliate at USC’s Center for Computational and Experimental Genomics, and the sole researcher there in plant genetics.