Machine learning reveals “important genes” in agriculture and medicine

Ad Blocker Detected

Our website is made possible by displaying online advertisements to our visitors. Please consider supporting us by disabling your ad blocker.

Corn grown in the NYU Roseson Zeger greenhouse on the roof of the NYU Genomics Center. Systems biology. Photo credit: NYU Coruzzi Lab

Machine learning can identify “important genes” that help grow plants with less fertilizer, Nature Communications.

Using Genomic Data Predicting agricultural and medical outcomes is systems biology. Researchers have worked to figure out how to best use the vast amount of available genomic data to predict how organisms will respond to changes in nutrient, toxin and pathogen exposure. Rice field. However, accurately predicting such complex outcomes in agriculture and medicine from information on the genomic scale remains an important challenge.

In Nature Communications’ study, researchers and staff at New York University in the United States and Taiwan tackled this challenge with machine learning, a type of artificial intelligence used to discover patterns in data.

“Focusing on genes whose expression patterns are evolutionarily conserved between species improves the ability to learn and predict“ important genes ”in the growth potential of staple foods and the consequences of animal diseases. It shows, ”stated Gloria Corzzi, Carol & Milton. Professor Petrie of the School of Biological Sciences and the Center for Genomics and Systems Biology at New York University and lead author of the article.

“Our approach takes advantage of natural phenotypic changes associated with genome-wide expression within and between species,” said New York University’s Center for Genomics and Systems Biology and lead author of the study. Chia-Yi Cheng from National Taiwan University added. “Reducing genomic input for genes whose expression patterns are conserved within and between species is a biologically principled way to reduce the dimension of genomic data, and machine learning models identify important genes. It shows that it greatly improves the ability to do. Properties. “

Corn grown in NYU’s Roseson Zeger greenhouse on the roof of the NYU Genomics Center. Systems biology. Photo credit: NYU Coruzzi Lab

As a proof of concept, researchers have demonstrated genes whose nitrogen reaction is evolutionarily conserved between two different plant species. Arabidopsis is a small flowering plant that is commonly used as Arabidopsis. Model organism Plant biology and corn varieties, America’s largest crop, have significantly improved the ability of machine learning models to predict key genes about how efficiently plants use nitrogen. Nitrogen is an important nutrient for plants and the main component of fertilizers. Plants that use nitrogen more efficiently grow better and require less fertilizer, which offers economic and environmental benefits.

The researchers conducted experiments to verify eight master transcription factors as genes that are important for nitrogen use efficiency. They have shown that altered gene expression in Arabidopsis and corn can promote plant growth in nitrogen-poor soils and have been tested in both NYU laboratories and the University of Illinois corn fields.

“We can now more accurately predict which maize hybrids use nitrogen fertilizers better in the fields, so that we can quickly improve this property. Nitrogen Use Efficiency Research author Stephen Moose, professor of plant science at the University of Illinois at Urbana-Champaign, said:

In addition, researchers have shown that this was evolutionary. Machine Learning This approach can be applied to other traits and species by predicting additional plant traits such as the biomass and yield of Arabidopsis and corn. They also showed that by studying mouse models, this approach can predict genes that are important for drought tolerance in another staple crop, rice, and the consequences of animal diseases.

“We have shown that our evolutionarily informed pipeline can also be applied to animals, so this underscores the possibility of uncovering them. It is important for the physiological or clinical properties of interest in biology, agriculture or medicine in general, ”says Coruzzi.

“Many important traits of agricultural or clinical importance are genetically complex, making it difficult to identify their control and inheritance. Our success lies in big data and thinking at the system level. It proves that it can handle the infamous and difficult task more easily, ”says research author Professor Ying Li. From the Faculty of Horticulture and Landscape Architecture at Purdue University.

Increase plant biomass: Biologists discover the molecular connection between nutrient availability and growth

For more informations:
Evolutionarily informed machine learning strengthens the power of predictive relationships between genes and phenotypes. Nature communication (2021). DOI: 10.1038 / s41467-021-25893-w

Provided by
New York University

Quote: Machine learning, agriculture and medicine, acquired on September 24, 2021 from (September 24, 2021) So) reveals ” important genes ”

This document is subject to copyright. No part may be reproduced without written permission except in fair transaction for personal investigation or research purposes. The content is provided for informational purposes only.

Source link Machine learning reveals “important genes” in agriculture and medicine