Rapid Phenotyping of Seed Oil Content

Abstract

Source: Adobe Stock. While there are great advances in crop genotyping, many research programs still depend upon the selection of plants or seeds based on the phenotype. This can involve sorting through thousands of samples by hand. Finding ways to automate sorting based on phenotype can increase the productivity of plant breeders, freeing up time to do other tasks. Albrecht Melchinger, ASA and CSSA member and Professor of Applied Genetics and Plant Breeding at the University of Hohenheim in Germany, uses seed phenotyping in his work. One example is the use of a color marker, “where you can see from the embryo coloration whether it's a haploid seed or a diploid seed. The haploid seeds are, in this case, white and the other ones are purple.” However, some germplasm is naturally purple and could not be used in the breeding program. To solve this problem, Melchinger and colleagues developed an inducer with high oil content. Haploid seeds would have normal oil content while the diploid seed had higher oil content. While color was no longer limiting the germplasm that could be used, researchers still had to perform the time-consuming task of analyzing individual seeds to determine oil content. To speed this process, these researchers have created a platform for determining oil content. Although they were working with maize, they realized this automated, high-throughput system had the potential to benefit breeders working with other oil crops. Oil crop breeders are often trying to increase oil content, “and it would be very nice if you had measurements of individual seeds in a nondestructive manner,” Melchinger says. An article recently published in Crop Science (http://bit.ly/2FLCqAX) describes this phenotyping platform for measuring the oil content of seeds and tests accuracy across a range of oil crops. Researchers used the platform to measure the oil content of canola, castor bean, cotton, jatropha, maize, soy, and sunflower. The platform has four modules (Fig. 1). The first separates individual seeds from a larger sample using suction. Depending upon the size and shape of the seed, the pneumatic pressure required to select a single seed needs to be adjusted. The second module determines seed mass. Mass is measured on a balance, and it is key to keep this clean and free from debris. Oil mass is measured in the third module using commercial TD-NMR (time domain nuclear magnetic resonance) equipment. A computer then calculates oil content from oil mass and seed mass data. This step is also one that needs to be adjusted based on seed size. Flow chart of the seeds through the modules of the high-throughput platform. NMR, nuclear magnetic resonance. The final module sorts seeds based on oil content, which can be done in two different ways. One approach is to set categories. For example, when sorting based on oil content to separate haploid and diploid seeds, a user can sort seeds into two categories. Alternatively, the module will measure each seed and set them on a tray in a grid pattern. The computer tracks the placement of each seed. A researcher can then query the dataset, for example identifying the top 10% of seeds based on oil content. Seeds meeting the selected criteria are identified by LED lights, which are located under each seed. In testing the system with these seven oil crops, the researchers report that their high-throughput phenotyping platform has high accuracy. Because the process is fully automated, a user can load seeds for analysis and walk away. “We do it very often overnight,” Melchinger says. He explains the pneumatic system that moves seeds through the modules is the key development, and the researchers have applied for a patent on this technology. They are also developing a manual that will outline how settings should be adjusted based on seed size and shape when using this platform for different crops. Melchinger sees this platform being useful beyond measuring oil content. “Our system makes use of NMR, but it is not confined to NMR,” he says. As technology is developed to analyze other traits, there is potential to switch the components while maintaining the high-throughput functionality of the platform. Slide 1: Module 1: A feeder (or hopper) is filled with seeds and the separator uses pneumatic pressure to select a single seed at a time for processing. Slide 2: Module 2: Seeds are weighed on a mass balance. Slide 3: Module 3: The NMR machine, where oil mass is determined. Seeds are transported into and out of the NMR machine using pneumatic pressure. Slide 4: Module 4a: Seed sorting based on pre-established categories. Seeds can be sorted into as many as six defined categories. Slide 5: Module 4b: Seed sorting onto a tray. Data for each seed are stored in the computer, and a user can define criteria for selection. LED lights under tray identify seeds that meet the criteria. Slide 6: LED selection grid in Module 4b: Seeds with light shining below meet user-defined selection criteria. Check out the Crop Science article, “High-Throughput Precision Phenotyping of the Oil Content of Single Seeds of Various Oilseed Crops” at: http://bit.ly/2FLCqAX.

Plant phenotyping relevance

種子油含量を個別・非破壊・高スループットで測定および選別するプラットフォームの開発、精度評価、複数作物への適用が中心であり、植物フェノタイピング手法に該当する。

Code and data availability

公開論文であることは確認できましたが、現在の公式API・許可済み取得経路では本文を自動取得できませんでした。

No evidence-backed public reproduction asset is currently recorded.