Plant scientists were able to augment corn genes identified by the new algorithm, effectively boosting nitrogen uptake and improving plant growth in nitrogen-poor soils. Photo by NYU Coruzzi Lab
Sept. 24 (UPI) -- Humans and plants have thousands of genes. Traditionally, studying the function of a single gene or group of genes required extensive experimentation.
Computers and access to large databases of genomic data, however, allow researchers to study gene functionality more efficiently. Still, mining massive amounts of genomic data is difficult for even the most powerful computer.
In a new breakthrough, researchers in the United States and Taiwan have developed a machine learning algorithm to more efficiently identify "genes of importance" in agriculture and medicine.
The machine learning algorithm, described Friday in the journal Nature Communications, could help scientists better anticipate how plants and animals will respond to changes in nutrition, toxins or pathogens -- allowing researchers to develop more resilient crops, diagnose rare diseases or anticipate the next pandemic.
"We show that focusing on genes whose expression patterns are evolutionarily conserved across species enhances our ability to learn and predict 'genes of importance' to growth performance for staple crops, as well as disease outcomes in animals," senior study author Gloria Coruzzi, biology professor at New York University's Center for Genomics and Systems Biology, said in a news release.
Essentially, researchers found a way to reduce the genetic noise to which an algorithm is subjected.
"We show that paring down our genomic input to genes whose expression patterns are conserved within and across species is a biologically principled way to reduce dimensionality of the genomic data, which significantly improves the ability of our machine learning models to identify which genes are important to a trait," said lead author Chia-Yi Cheng, researcher with the Center for Genomics and Systems Biology and National Taiwan University.
In a proof-of-concept experiment, researchers showed that nitrogen-reactive genes are evolutionarily conserved between two diverse plant species: Arabidopsis, a small flowering plant and popular plant model among plant scientists, and several varieties of corn. With noise from the input data reduced, the new algorithm efficiently and successfully identified genes important for nitrogen processing.
Nitrogen uptake is essential for plant growth. Plants engineered to more efficiently absorb and use nitrogen could reduce fertilizer use. Nitrogen overuse has been implicated in a variety of environmental problems, including nutrient loading, harmful algae blooms and coral bleaching.
In follow-up experiments, researchers confirmed the importance of the genes identified by their algorithm. The plant scientists were able to augment the genes of corn varieties to boost nitrogen uptake and improve plant growth in nitrogen-poor soils.
"Now that we can more accurately predict which corn hybrids are better at using nitrogen fertilizer in the field, we can rapidly improve this trait," said co-author Stephen Moose, professor of crop sciences at the University of Illinois at Urbana-Champaign. "Increasing nitrogen use efficiency in corn and other crops offers three key benefits by lowering farmer costs, reducing environmental pollution, and mitigating greenhouse gas emissions from agriculture."
In addition to identifying genes relevant to various crop traits, researchers suggest their algorithm can be used to anticipate genes relevant to disease outcomes in mouse models, which could inspire the development of new therapies and diagnostic techniques.
"Because we showed that our evolutionarily informed pipeline can also be applied in animals, this underlines its potential to uncover genes of importance for any physiological or clinical traits of interest across biology, agriculture, or medicine," said Coruzzi.