Data mining in agriculture has the potential to be extremely big business. As Dr James Hane explains, the approach is leading researchers to a better understanding of crop protection.
Dr James Hane, from the Centre for Crop and Disease Management, uses bioinformatics to analyse how fungal pathogens infect and adapt.
“Certain fungal genes produce proteins called effectors which suppress plant immunity, promote disease symptoms, and ultimately cause cell death,” explains Hane. “But if you can isolate the effector proteins, you can test them against a range of crop cultivars and rapidly determine whether they will be resistant or susceptible to the fungus.”
First, you need to identify the handful of effector genes among the several thousand genes encoded by a fungal genome. Whole-genome sequencing has massively sped up this process. In the mid-2000s it cost several million dollars to sequence a fungal genome, so only a few important species were sequenced – now it costs less than a thousand dollars.
“We’re moving from studying a single reference genome for a single species, to comparing hundreds of genomes of the same species. Comparative genomics can then pinpoint DNA regions with abnormal mutation levels. When the host crop is developing resistance, there’s pressure for the pathogen to evolve, so sites of higher variability between genomes may suggest that we’ve found an effector gene,” says Hane.
“We have also developed a range of other techniques that can sift through huge amounts of genome sequence data to quickly identify genes likely to cause plant disease.” The aim is to fast-track effector gene information to molecular biologists for testing, and ultimately to growers, allowing informed decisions about which cultivars to grow to maximise disease resistance.
“We’re now extending our comparative genomics approach to comparing geographic regions. Spatially mapping the distribution of fungal strains, effectors and other useful data such as fungicide resistance, will allow growers to choose cultivars and fungicide applications tailored to their region, providing optimal resistance at the lowest cost.”
The next step is to move beyond the isolated consideration of a single species at a time. Metagenomics studies DNA from all species present in an environmental sample – not just the pathogen of interest. Next-generation sequencing ‘reads’ millions of short DNA sequences in these mixed samples, making it possible to not just identify, but also measure relative abundances of microbial species present in agricultural soils or infected plant samples.
Metagenomic analysis across regions and seasons will allow Hane to map the changing distribution of pathogens, effectors and resistance, and identify patterns where species co-occur. This will help growers decide which cultivars to grow, which fungicides to apply (and when), and even which microbial populations may be worth introducing into their soils.