AMES, Iowa : In its native habitat, switchgrass flowered earlier when growing farther north. In experiments with diverse genetic samples, it flowered earlier in the south.
The discrepancy wasn’t a welcome sight for a research team studying how prairie grasses respond in different environments, but resolving the apparent conflict led the scientists to draw better conclusions, showing the potential value of large public data sets collected in natural growing conditions.
The study published late last month in the high-profile, peer-reviewed scientific journal Cell, “Harnessing citizen science to contextualize adaptation mechanism discovery,” integrated trends culled from AI-powered scans of tens of thousands of online photos of perennial grasses with findings from two years of growing switchgrass at research sites across the Midwest and Gulf regions as well as a detailed molecular characterization of three underlying genes involved in switchgrass flowering.
Led in part by Iowa State University agronomy professor Jianming Yu, the research team identified the genetic basis for adaptative responses that help explain the contradictory flowering-time trends. But the framework wringing useful insight from “citizen science,” in this case an online cache of photos mostly from iNaturalist – is as notable as the findings, said Yu, the Pioneer Hi-Bred Distinguished Chair in Maize Breeding and director of the Raymond F. Baker Center for Plant Breeding.
“With this study, we have connected our quantitative genetic and genomic research with ecology, evolution and adaptation over a large-scale landscape. The beauty is we’re bridging them together so we can see the whole picture,” he said.
Conflicting data
Researchers built an AI tool to screen nearly 44,000 photos of warm-season grasses with time and location data, yielding about 5,000 observations of flowering switchgrass, big and little bluestem, and indiangrass. In each of the four species, the average flowering time was earlier in the north than the south.
Focusing on switchgrass, the most studied of the four grasses, researchers analyzed data from genetic mapping populations of switchgrass to identify a gene network associated with flowering time. Studying a diversity panel collected as samples from wild-grown switchgrass, they found three haplotypes – combinations of variants of the three underlying genes linked to flowering. Each haplotype was primarily found in geographic clusters, including a variant specific to the Midwest and one common in Gulf Coast states.
Switchgrass samples containing haplotypes associated with the Midwest and Gulf regions were grown at 10 research gardens for two years, flowering on average 2.3 days later for every degree of latitude farther north – opposite of the data from native habitats.
“We had to wrap our minds around that,” Yu said.
Adaptation at work
Added context pointed a way toward a solution. Analyzing the expected flowering time of the diversity panel with a model that included genetic and environmental data showed the temperature from April 25 to May 5 had the strongest correlation with flowering time. Warm weather during that period sped up flowering by 3.4 days for each degree Celsius.
But the switchgrass haplotype common in the southern Gulf region, referred to in the paper as H1, tended to flower 45 days later in all instances than the Midwestern haplotype, H2. And H2 flowering was more sensitive to temperature during the critical time in late April and early May, the researchers found.
The differences between H1 and H2 make sense for their respective geographies, Yu said. In the north, H2 sprouts flowers earlier because the risk of extreme heat in the summer is lower, but temperatures can get cold in the fall. There’s an advantage to flowering earlier so switchgrass can turn its attention to preparing for the winter.
Farther south, H1’s delayed flowering helps it avoid reproducing during the height of summer, and holding off until late summer poses less risk because fall temperatures are typically more moderate than in the north.
“In their native conditions, both haplotypes are doing the things that they need to do to survive and thrive,” Yu said. “In the north, they flower earlier because winter is coming. But in the south, there’s no rush because summer is so hot and the fall is mild.”
A powerful pairing
Yu said the work shows the effectiveness of pairing controlled research and natural data sets, especially in studying how plants adapt in different environments – a concept also known as phenotypic plasticity.
Without considering the public database of photos, researchers wouldn’t have spotted the flowering adaptation in native habitats. Without studying genotyped plants, they wouldn’t have been able to understand it.
“Our study highlights the power of combining citizen science observations with designed experiments to uncover mechanisms of adaptation across spatiotemporal scales,” wrote the study’s 27 co-authors, including two scientists from the U.S. Department of Agriculture – senior author Xianran Li and first author Laura Tibbs-Cortes – with a history at Iowa State, Li as a research associate professor and Tibbs-Cortes as a doctoral student.
“It was the collective effort by scientists across a wide range of disciplines and institutions, led by the senior author Li, that gathered all the evidence to assemble the puzzle. We hope it inspires other studies,” Yu said.
While relevant publicly collected data isn’t always available, it should be integrated with plant experiment data when it is, Yu said.
“You can’t say, ‘No, the experiment is true,’ and just ignore citizen science,” he said. “You’ve got to put them together.”
