Progress 06/15/18 to 10/14/20
Outputs
Target Audience:The target audienceof the technical aspects of this project include the plant phenotyping community other plant scientists includinggeneticists (particularly maize and other cereal researchers), physiologists, developmental biologists,biometricians, and breeders. This audience was reached through several scientific conferencesand through a peer-reviewed publication, with additional publications in progress.Additionally, outreach efforts for the duration of this project included local students at La Salle Middle School, a student mentee at Fontbonne Hall Academy, an NSF REU student associated with the laboratory,graduate students at Washington University in St. Louis, participants at a workshop at the Cold Spring Harbor Laboratories,and local public attendees of theSaint Louis Science Center's SciFest Bright Ideas Expoand theDonald Danforth Plant Science Center'sRaspberry Pi Jam Festival.More specifically, during the 2018-2019 reporting period, weekly after-school educationalsessions were held to teach La Salle students a variety of plant-related topics, including hands-on activities. Furthermore, a high school student at Fontbonne Hall Academy was given technical guidance on how to use public cyberinfrastructure resources to analyze plant root images, and how to then analyze the data using a statistical programming language, ultimately participating in a regional scientific competition held by Regeneron. Along similar lines, an NSF REU student in the laboratory was given technical guidance on analyzing root hydropatterning data. Beginning in this period and extending into the 2019-2020 reporting period, first-year graduate students at the Washington University in St. Louis were also provided technical instruction on image analysis and data analysis in a series of guest lectures. This lesson plan was also used as part of a workshop on cereal genomics at the Cold Spring Harbor Laboratories. Finally, participating in single-day events included illustrating the use of new technologies in plant sciencesat the Saint Louis Science Center and the Raspberry Pi Jam Festival at the Donald Danforth Plant Science Center. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?Technical development provided through the project including: direct supervision and guidance under Chris Topp (mentor), who is a leadingexpertin plant phenotyping/imaging and genetics; hands-on experience and use of an X-ray computed tomography imaging system; hands-on experience with processing of 3D images; statistical analysis of 3D image data output; and finally opportunities for thetranslation of high-dimensional phenotyping data for biological studies withinthe context of genetics.Additionalprofessional development provided through the project include the opportunities for oral presentations and networking through attendance of scientific conferences; pedagogy training through classes and workshops taken at the Washing University in St. Louis; opportunities for outreach activities;and experiences in scientific writing through the publication of a peer-reviewed article, with another peer-reviewed article in submission, and a scientific review article in progress. How have the results been disseminated to communities of interest?The results have been disseminated to communities of interest through an already-published peer-reviewed publication,four scientific conference presentations given during the project lifetime, and an additional peer-reviewed publication in submission and review article in progress. Additionally, general information on plant biology, especially regarding roots and imaging, have been disseminated to students and the public through outreach activities conducted during the project lifetime What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
Crossesbetween perennial teosinte (Zea diploperennis, PI 462368) and 4 other Zea (sub)species have been made: Zea mays cv. B73, Zea mays sp. mexicana (PI 566673), Zea mays sp. parviglumis (PI 658213), and Zea luxurians (PI 207452); in all cases, Zea diploperennis was used as the female. We established a workflow for the imaging and segmentation (digital isolation) of roots in potting media using X-ray computed tomography. However, imaging of rhizomes using positron emission tomography in addition to X-ray computed tomography, including with registration (alignment of two imaging types), has not yet been achieved with this germplasm. Nevertheless, we have established a protocol for root and subterranean plant phenotyping, and have 3 systems for digitally isolating the root system: the first two methods involve traditional thresholding methods using commercial and academic software (with different workflows depending on whether the roots are imaged within soil, or excavated and washed first), while the third method involves a convolutional neural network approach to segment roots from their growth substrates have been developed using maize roots as training sets. From all methods, the end result is digital 3D reconstruction of the root system amenable to analysis. This workflow has been successfully used on maize and sorghum roots, and the resulting root models can be processed and measured. From this pipeline, over 70 3D root traits can be quantified from X-ray computed tomography imaging, with high broad-sense heritabilities demonstrating an improvement over 2D based imaging. Using multivariate supervised classification methods such as linear discriminated analysis and random forest, root systems of various genotypes or growth conditions can be distinguished from one another and the most influential root traits contributing to their differences can be found. Additional refinements have been made in downstream biometrical analysis, including accounting for how tillering affects root architecture traits and quantification. Finally, through external collaborations we now have developedanother root-based image analysis tool that can estimate over 60 additional traits based explicitly on the topology of the root system, e.g. specific measurements on different hierarchies of the root, which allows for example the automated measurements of primary versus lateral roots. Such topological features are novel in 3D root system phenotyping, and represent a significant advance in the field.
Publications
- Type:
Journal Articles
Status:
Submitted
Year Published:
2021
Citation:
Shao MR, Jiang N, Li M, Howard A, Lehner K, Mullen JL, Gunn S, McKay JK, Topp CN. Complementary Phenotyping of Maize Root System Architecture by Root Pulling Force and X-Ray Computed Tomography. Plant Physiology.
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2018
Citation:
3D phenotyping of root crown and inflorescence architecture across diverse genetics in maize and sorghum with X-Ray CT. The 5th International Plant Phenotyping Symposium.
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2019
Citation:
High-resolution 3D phenotyping of diverse maize root and sorghum inflorescence architectures. XXIVth EUCARPIA Maize and Sorghum Conference.
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2019
Citation:
Quantifying Maize Root-Shoot Plasticity and 3D Architectural Changes from Water Stress using Precision Phenotyping. The 61st Annual Maize Genetics Conference.
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2020
Citation:
Comprehensive 3D Phenotyping of Inflorescence and Root Architectural Variation in Sorghum. The 4th Annual Phenome 2020 Conference.
|
Progress 06/15/19 to 06/14/20
Outputs
Target Audience:During this reporting period, formal classroom instruction was given to graduate students at the Washington University in St. Louis, as part of a session on plant phenotype, imaging, and data analysis. Students were taught on the use of ImageJ to segment and threshold images of plant leaves, measure their shapes, then analyze multivariatedata using R programming language. A similar lesson plan was used to teach part of the "Workshop on Cereal Genomics"at Cold Spring Harbor Labs.In additional, individual mentoring of a student at Fontbonne Hall Academy continued and was eventually concluded with the end of the research project, which the student submitted to thenation-wide Regeneron science competition. Finally, two scientific conference were attended where scientific communication and exchanges could be made: The 2019EUCARPIA Maize and Sorghum Conference, and the 2020 Phenome Conference. Changes/Problems:Imaging by positron emission tomography (PET) for guiding rhizome assays has not been successful and will not be pursued further. Instead, focus will be given on gene expression of the interspecific hybrids between annual maize/teosinte and perennial teosinte. What opportunities for training and professional development has the project provided? This project has provided training activities through one-on-one work with Chris Topp (mentor) and other technical experts that facilitated increased technical proficiency with X-ray computed tomography. This includes active hands-on use of the North Star Imaging X5000 system, subsequent image processing, root quantification, statistical analysis, and visualiation. Discussions and troubleshooting related to scanning, segmenting, and thresholding of rootsare conducted by weekly meetings related to X-ray computed tomography imaging. This experience and expertise in 3D plant imaging will be the basis of future career steps. This project has also provided professional development through attendance and networking within two international scientific conferences. Pedagogy experiencewas alsogained through teaching at Washington University in St. Louis and the Cold Spring Harbor Labs. Finally, an invitation to author a review article in the journalPlant Physiology was accepted and is currently being drafted, per the goals of the project proposal,with an anticipated publication date in 2021. How have the results been disseminated to communities of interest?Research on grass roots was presented in two international conference within the reporting period, the 2019 EUCARPIA Maize and Sorghum Conference and the 2020 Phenome Conference,in each case delivered through a talk to 100+ scientists. Additionally, amaize root research project undertaken by a mentoree (Fontbonne Hall Academy) was submited to a national competition where it was awarded a finalist position within the environmental science category. What do you plan to do during the next reporting period to accomplish the goals?The no-cost extension period awarded to the project will enable thecompletion of the RNA-seq experiments, particularly with respect to the annual x perennial teosinte hybrids and the teosinte x maize hybrids. The additional time will also allow further X-ray computed tomography imagingand analysis of perennial grass root systems.
Impacts
What was accomplished under these goals?
Multiple interspecific crosses between maize and teosinte have been generated, producing valuable material for examining gene expression changes in these hybrids. These will now be the main focus on gene expression experiments, as to-date positron emission tomography imaging with teosinte rhizomes has not been successful.With respect to root phenotyping,additional refinements have been made to theanalysis based on custom software tools and scripts, including how to extend the downstream analysis to account for how tillering affects root architecture traits and quantification. Currently, over 70 3D root traits can be quantified from X-ray computed tomography imaging, with high broad-sense heritabilities demonstrating an improvement over 2D based imaging.Using multivariate supervised classification methods such as linear discriminated analysis and random forest, root systems of various genotypes can be distinguished from one another and the most influential root traits contributing to their differences can be found. Additionally, convolutional neural network approaches to segment roots from their growth substrateshave been development using maize roots as training sets. These can be applied to the root systems of other species to accelerate and improve the accuracy of their segmentation and quantification.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2020
Citation:
Li M, Shao MR, Zeng D, Ju T, Kellogg EA, Topp CN. Comprehensive 3D Phenotyping reveals Continuous Morphological Variation across Genetically Diverse Sorghum Inflorescences. New Phytologist. 2020 Jun;226(6):1873-1885. doi: 10.1111/nph.16533.
|
Progress 06/15/18 to 06/14/19
Outputs
Target Audience:During this reporting period, outreach efforts were performed towards target audiences including the local public, graduate students,high school students, anddisadvantaged middle school students. Major outreachincluded participation in a weeklyafter-school plant science enrichment program at La Salle Middle School in Saint Louis, MO. As part of this program, approximately one dozen disadvantagedstudents were provided educational information on a variety of topics including root function, photosynthesis, nutrients, and more. Each session also included hands-on activities such as microscopy use, root anatomy, and demonstrations of various physical and biological properties such as diffusion, water transport, transpiration, more. Another local outreach effortwas participation in the Saint Louis Science Center's SciFest Bright Ideas Expo, where public attendees were given demonstrations on root architecture from excavated samples and from virtual reality imaging. Within the Donald Danforth Plant Science Center, outreach efforts included participation in the Raspberry Pi Jam Festival, where youth were taught basics ofsoldering and circuitry,and the organization of BioBash, anannual event hosting a career discussion panel and a networking session. Individual mentoring included direct co-instruction of an NSF REU student, who was guided on using R programming language and statistics for analyzing lateral root formation under hydropatterning, and the online-based mentoring of a high school student at Fontbonne Hall Academy, who was instructed on how to use public cyberinfrastructure resources for the quantitative analysis of maize root images. Finally, formal classroom instruction was given to graduate students at the Washington University in St. Louis,as part of a session on plant phenotype, imaging, and data analysis. Students were taught on the use of ImageJ to segment and threshold images of plant leaves, measure their shapes, then analyze the data using R programming language. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?This project has provided training activities through one-on-one work with Chris Topp (mentor) and other technical experts that facilitated increased technical proficiency with X-ray computed tomography. This includes active hands-on use of the North Star Imaging X5000 system, and subsequent image processing and root quantification. Several dozen root crowns were imaged during this training, followed by analysis using custom software tools and scripts. Discussions and troubleshooting related toscanning, segmenting, and thresholding ofroots in pots containing soil media was also conducted throughweekly meetings on X-ray computed tomography imaging. This project has also provided professional development through attendance of the Maize Conference in St. Louis, a prestigious conference for maize research, with dozens of speakers and hundreds of posters. How have the results been disseminated to communities of interest?The importance of underground plant biology was desseminated to the public at the SciFest Bright Ideas Expo at the St. Louis Science Center, where attendees were able to hold and examine root crowns of various maize genotypes, and explore a maize root crown in full 3D through use of a virtual reality headset. Additionally, participants would listen to explanations about root architecture and biology during this experience. Furthermore, root biology was a key topic taught to students at the La Salle Middle School outreach event, including the function of roots, root anatomy, root types, and root architecture. What do you plan to do during the next reporting period to accomplish the goals?During the next reporting period, plans to accomplish the project goals and objectives include: Extraction of RNA fromF1 plants and parental plants (generated in current reporting period), followed by high-throughput Illumina sequencing and analysis. Metabolic and transcriptomic profiling of active rhizomes, guided by registered positron emission tomography and X-ray computed tomography images. X-ray imagingand quantification of roots from additional rhizatomous grass species for comparative analysis, using workflows and conditions established during the previous reporting period usingmaize root systems. Construction of a hydroponic system to facilitate observation of root and rhizome development in these various rhizomatous grass species, to better anticipate ideal imaging window for plants grown in soil potting media.
Impacts
What was accomplished under these goals?
The impact of this project, if successful, is expected to provide significant towards understanding how rhizomes form and are controlled in wild perennial relatives of maize (aka corn). These wild perennial relatives, also known as perennial teosinte,are able to re-grow for multiple yearsfrom suchunderground rhizomes.Undestanding how this occurs, and what are the genetic barriers preventing this from beingbredinto domesticated corn, is the first step towards creating perennial varieties of domesticated corn. Such varieties would require less water and fertilizer inputs, resulting in increased agricultural sustainability and decreased pollution from runoff. Additionally, use of perennial corn would be expected to enhance soil health and organic matter, increase carbon sequestration, and decrease erosion levels.In order to achieve this, we are utilizing recent advances in imaging,gene expression analysis, and metabolic profiling to understand the physiology of maize relativesand other grasses which can form rhizomes, compared to the maizespecies and subspecies which cannot. Progress on project goals: 1) Integrated transcriptome and phytohormone profiling of rhizome development and growth in perennial teosinte. Prior to transcriptome and phytohormone profiling, imaging of rhizomes using X-ray computed tomography and positron emission tomography must be established and registration (alignment of two imaging types) performed. We have established a workflow for the imaging and segmentation (digital isolation)of roots in potting media(see Objective 3 below). We are also continuing to optimize the results and registration of positron emission tomography imaging. Once sufficient optimization is achieved, dissection to obtain samples for transcriptome and phytohormone profiling can occur. 2) Identification of allelic expression biases and non-additive expression in perennial teosinte F1 hybrids. Multiple crosses between perennial teosinte (Zea diploperennis, PI 462368) and 4 other Zea(sub)species have been made: Zea mays cv. B73, Zea mays sp. mexicana(PI 566673), Zea mayssp.parviglumis (PI 658213), andZea luxurians (PI 207452). Parental plants were grown under short-day conditions to trigger flowering. In all cases, Zea diploperennis was used as the female. Multiple individuals from each cross and parental genotype will be extracted for RNA and subjected to RNA-seq for analysis. 3) Comparative physiology of grass rhizomes using X-ray computed tomography. Successfuly X-ray computed tomography of root systems in soil is technically challenging. Currently, we have established a protocol for acceptable scanning parameters using the North Star Imaging X5000, and have 3 systems of digitally isolating the root components from the soil: the first two methods involve traditional thresholding methods using commercial and academic software, while the third method involves a custom machine-learning approach whereby a model is trained based on a proportion of the overall images, which then can be used to process thousands of images. From all methods, the end result is digital 3D reconstruction of the root system amenable to analysis. This workflow has been successfully used on maize roots, and the resulting root models can be processed and quantified using existing custom software.We are currently cultivating multiple perennial species of grasses in the greenhouse, and the next step is to image these root systems.
Publications
|