Scientists stack algorithms to improve predictions of yield-boosting crop traits

CHAMPAIGN (June 3, 2019) — Hyperspectral data comprises the full light spectrum; this dataset of continuous spectral information has many applications from understanding the health of the Great Barrier Reef to picking out more productive crop cultivars. To help researchers better predict high-yielding crop traits, a team from the University of Illinois have stacked together six high-powered, machine learning algorithms that are used to interpret hyperspectral data—and they demonstrated that this technique improved the predictive power of a recent study by up to 15 percent, compared to using just one algorithm.

“We are empowering scientists from many fields, who are not necessarily experts in computational analysis, to translate their enormous datasets into beneficial results,” said first author Peng Fu, a postdoctoral researcher at Illinois, who led this work for a research project called Realizing Increased Photosynthetic Efficiency (RIPE). “Now scientists do not need to scratch their heads to figure out which machine learning algorithms to use; they can apply six or more algorithms—for the price of one—to make more accurate predictions.”

RIPE, which is led by Illinois, is engineering crops to be more productive by improving photosynthesis, the natural process all plants use to convert sunlight into energy and yields. RIPE is supported by the Bill & Melinda Gates Foundation, the U.S. Foundation for Food and Agriculture Research (FFAR), and the U.K. Government’s Department for International Development (DFID).

In a recent study, published in Remote Sensing of Environment, the team introduced spectral analysis as a means to quickly identify photosynthetic improvements that could increase yields. In this new study, published in Frontiers in Plant Science, the team improved their previous predictions of photosynthetic capacity by as much as 15 percent using machine learning, where computers automatically applied these six algorithms to their dataset without human help.

“I’ve loved seeing what’s possible when you can use computational power to exploit the data for all its worth,” said co-author Katherine Meacham-Hensold, a RIPE postdoctoral researcher at Illinois, who led the previous study in Remote Sensing of Environment. “It’s exciting to see what a data analyst like Peng can do with my data. Now other non-data-analyst scientists can test several powerful algorithms to figure out which one will help them leverage their data to the fullest extent.”

However, more studies are needed to prove the relevance of this stacked algorithm technique to the plant science community and other fields of study.

“By applying the expertise of data analysts to address the needs of plant physiologists like myself, we ended up refining a technique that is relevant to other hyperspectral datasets,” said co-author Carl Bernacchi, a RIPE research leader and scientist with the U.S. Department of Agriculture, Agricultural Research Service, who is based at Illinois’ Carl R. Woese Institute for Genomic Biology. “The next step is to test more stacked machine learning algorithms on datasets from many more crop species and explore the utility of this technique to estimate other parameters, such as abiotic stresses from drought or disease.”

“As scientists, we should try to use our domain knowledge to explain advanced performance from machine learning methods,” said co-author Kaiyu Guan, an assistant professor in Illinois’ College of Agriculture, Consumer, and Environmental Sciences (ACES). “Combining computational methods and domain disciplines allows us to possibly unravel what causes the measurable differences in hyperspectral datasets—which is an unsolved mystery in our work and worth future exploration.”

###

Realizing Increased Photosynthetic Efficiency (RIPE) is engineering staple food crops to more efficiently turn the sun’s energy into food to sustainably increase worldwide food productivity, with support from the Bill & Melinda Gates Foundation, the U.S. Foundation for Food and Agriculture Research, and the U.K. Government’s Department for International Development.

RIPE is led by the University of Illinois in partnership with The Australian National University, Chinese Academy of Sciences, Commonwealth Scientific and Industrial Research Organisation, Lancaster University, Louisiana State University, University of California, Berkeley, University of Essex, and U.S. Department of Agriculture, Agricultural Research Service.


Editor’s Notes:

The paper “Hyperspectral leaf reflectance as proxy for photosynthetic capacities: an ensemble approach based on multiple machine learning algorithms” published by the journal Frontiers in Plant Science is available online (DOI: 10.3389/fpls.2019.00730) or by request.

Media Contact:
Claire Benjamin
RIPE Communications Coordinator
claire@illinois.edu
+1-217-244-0941

Overcoming Water Scarcity

Overcoming Water Scarcity

Continue

Agriculture uses 70 percent of the world’s accessible freshwater. FFAR’s 2016-2018 Overcoming Water Scarcity Challenge Area addressed water use efficiency in agriculture by developing water conservation and reuse technologies, improving crop and livestock breeds, creating improved agronomic practices, increasing the social and economic tractability of conservation practices and enhancing the efficacy of Extension services.

FFAR’s Sustainable Water Management Challenge Area builds on earlier work to increase water availability and water efficiency for agricultural use, reduces agricultural water pollution and develops water reuse technologies.

Healthy Soils, Thriving Farms

Healthy Soils, Thriving Farms

Continue

FFAR’s 2016-2018 Healthy Soils, Thriving Farms Challenge Area increased soil health by building knowledge, fueling innovation, and enabling adoption of existing or new innovative practices that improve soil health.

The Soil Health Challenge Area advances existing research and identifies linkages between farm productivity and soil health, while also addressing barriers to the adoption of soil health practices.

Protein Challenge

Protein Challenge

Continue

FFAR’s 2016-2018 Protein Challenge Area sought to improve the environmental, economic and social sustainability of diverse proteins.

The Advance Animal Systems challenge area supports sustainable animal production through environmentally sound productions practices and advancement in animal health and welfare. Additionally, the Next Generation Crops Challenge Area develops non-traditional crops, including plant-based proteins, and creates new economic opportunities for conventional crops to increase future crop diversity and farm profitability.

Food Waste and Loss

Food Waste and Loss

Continue

About 40 percent of food in the US, or $161 billion each year, is lost or wasted. FFAR’s 2016-2018 Food and Waste Loss Challenge Area addressed the social, economic and environmental impacts from food waste and loss through research that developed of novel uses for agricultural waste, improved storage and distribution, supported tracking and monitoring, minimized spoilage through pre- and post-harvest innovations and changed behaviors to reduce food waste

FFAR’s current Health-Agriculture Nexus Challenge Area addresses food waste and loss and supports innovative, systems-level approaches to reduce food and nutritional insecurity and improve human health in the US and globally.

Forging the Innovation Pathway to Sustainability

Continue

Supporting innovation is necessary for sustainable results. Over the last 50 years, farmers have tripled global food production thanks to agricultural innovations. Forging the Innovation Pathway to Sustainability was a 2016-2018 Challenge Area that focused on understanding the barriers and processes that prevented the adoption of technology and research results into sustainable practices.

Urban Food Systems

Urban Food Systems

Continue

The 2016-2018 Urban Food Systems Challenge Area addressed feeding urban populations through urban and peri-urban agriculture and augmenting the capabilities of our current food system.

The Urban Food Systems Challenge Area continues this work and enhances our ability to feed urban populations.

Making My Plate Your Plate

Continue

FFAR’s 2016-2018 Making My Plate Your Plate Challenge Area focused on helping Americans meet the USDA 2015 Dietary Guideline recommendations for fruit and vegetable consumption, including research to both produce and provide access to nutritious fruits and vegetables.

FFAR’s current Health-Agriculture Nexus Challenge Area supports innovative, systems-level approaches to reduce food and nutritional insecurity and improve human health in the US and globally.