The weather may have been colder than expected for a Thursday morning in April, but that didn’t stop numerous agricultural students and professors from gathering in Throckmorton Hall to listen to Edward Buckler, plant research geneticist for the U.S. Department of Agriculture at Cornell University, give a lecture on some of the problems facing 21st century agriculture — and how he is trying to solve them.
According to Buckler’s website, Buckler Lab, one of the questions he and his team work to address is related to identifying harmful mutations in corn: how does genetic variation relate to phenotypic variation?
A phenotype refers to how a set of genes — known as a genotype — is physically expressed. Buckler works to see what genes cause certain traits to develop in fully grown crops and hopefully create more resilient produce from this knowledge.
Buckler began by describing a few different methods used for plant breeding over time: phenotypic selection, controlled trials, genomic selection and designed genomes. He called each of these breeding modes 1.0, 2.0, 3.0 and 4.0, respectively.
Buckler continued to speak about what his team was doing and why they were doing it, presenting a variety of graphs and data to showcase results and trends they have found in their research. Buckler’s group is also striving to increase crop productivity while remaining fiscally effective.
Buckler Lab mainly focuses on corn, otherwise known as maize, in its research. According to the lab’s website, maize is the world’s largest production crop and it has great potential for genetic modification.
Related:
Meet Sam Hankins, a local high school athlete with his sights set on the Olympics
As part of its focus on using genetics to combat challenges facing agriculture, Buckler Lab’s research falls into three general areas, according to its website. The first is called germplasm diversity, which involves analyzing molecular and phenotypic variations. The second is analysis and bioinformatic tools, looking at ways of identifying and measuring functional variations in species. Lastly, they find useful genetic variation for a range of different traits.
Buckler emphasized in his lecture that all of this information is helpful in determining how to create crops that can stand up to challenges like weather damage, insect populations and so on.
“Essentially, if every gene produces the right level of expression [and] it’s completely balanced, you get a nice rigorous plant,” Buckler said. “If you’ve got expression going all over the board there … you end up with a lousy plant.”
Buckler went into detail on how his team is using both evolutionary data and machine learning in order to find harmful — or “deleterious” — genes.
By identifying what and where these genes are on the genome, Buckler said it will be easier to engineer healthier and more productive crops, and they are striving toward this goal at Buckler Lab.
“The hypothesis that we’d really like to [pose] is that we could develop models that are based on the entire genome and based on the central dogma to help us identify deleterious mutations,” Buckler said.
The so-called central dogma of molecular biology is a way of describing the process by which genetic information flows into proteins, where DNA is transcribed to RNA and then translated to proteins. If mutation occurs in the DNA, the protein would subsequently be affected and could cause a variety of results.
Buckler said he thinks evolution and machine learning work well together when it comes to the future of agriculture.
“Evolution — its strength is this near infinite level of mutagenesis and the ability to let natural selection estimate the fitness effects in these various variants,” Buckler said. “And machine learning is allowing us to scale to sizes [where] learning can be generalized.”
