b'conditions. Direct selection for RSA has not occurred because the traits involved with it are complex, as is the measure-ment of these traits and the effects of environmental factors. In addition, there have been a large number of root traits identified through different studies and software programs have further com-plicated the identification of opportuni-ties to select the most informative and relevant suite of traits. Root traits are among the most dif-ficult to study in plant sciences, Singh explains. While root researchers have achieved success in trait identification, there are still challenges in the field-based root trait studies involving large number of varieties. Its hard to dig up field plants without disruption, and current sensors are not sensitive enough to do remote sensing in row crops. So, it requires a very commit-ted team to do long hours of work. Sourcing GeneticsTo begin the project, members of theManitobas Kevin Falk and Zaki Jubery and colleagues imaged thousands of soybean root team first reviewed the research onsystems and used machine to measure root traits and correlate differences in root system root morphology. They then developedarchitecture to the genome of the plants. Falk says the work will enable breeding for informative root trait categories whichdrought-tolerant root systems.go beyond individual root traits to describe differing root shapes and other broad characteristics.How This will Change Soybeantive for researchers to investigate if root The team chose 292 soybean varie- Breedingtraits have been optimized in commercial ties from the diverse soybean accessionsNow that this project has been published,varieties, or if there is a huge untapped held in the USDA core collection (primarilySingh believes it will begin to shift theresource of root trait variation out there. in maturity groups II and III) and from atraditional crop breeding focus fromWe also need to determine how subset of the soybean Nested Associationbeing solely focused on above-groundthe root traits influence yield and other Mapping parents. These accessions weretraits (easily observable and perceived asimportant traits, he says. Think of this as planted in controlled environment condi- sufficient to understand and predict plantmatching the chassis to the engine motor tions and the roots were imaged 6, 9 andperformance) to the whole plant.so both work in unison to deliver top and 12 days after germination. Computer visionFocusing on above-ground traits isconsistent speeds. and ML methods were used to measureeffective, but it completely misses theSeyed Vahid Mirnezami, who was part root traits and accessions with similarwhole other world of plant roots andof the project during his PhD studies and RSAs were categorized into eight groups.their anatomical, morphological andcurrently works in Bayers Crop Science Once the genetic and phenotypicphysiological traits, he says. Due to thedivision in Missouri, says ML and deep (observable) analysis was finished, thetechnical difficulties and resource inten- learning coupled with image processing team found there was genetic variabilitysiveness of observing root traits, we havetools could also be applied to measure for root shape, length, number, mass andnot put a lot of focus on them. But, withother crop traits, and above-ground traits angle. The accessions were geneticallycontinual innovations in imaging plat- could be correlated with the root traits. diverse but there were, as one wouldforms and computer vision/ML methods,Indeed, Singhs PhD student Clayton expect, group-specific trends.the research community can overcomeCarley is now working on this. The team also found that groupingsseveral of these challenges. The team is also conducting more based on genetic analysis correlated with Now that he and his colleagues havegenetic studies, building their under-geographical origins. In addition, its clearexposed the extent of genetic diversitystanding of the optimal combination that a large amount of the genotypesthat exists in important root traitsandof RSA traits and root nodulation, and (which represent cultivars originating infound that the majority of North Americanexploring the connections between the U.S.) lack genetic diversity for RSAsoybean varieties are limited in root traitcontrolled environment results and field-traits. diversitySingh believes its impera- based observations.SW110/ SEEDWORLD.COMDECEMBER 2020'