1
|
Berrigan EM, Wang L, Carrillo H, Echegoyen K, Kappes M, Torres J, Ai-Perreira A, McCoy E, Shane E, Copeland CD, Ragel L, Georgousakis C, Lee S, Reynolds D, Talgo A, Gonzalez J, Zhang L, Rajurkar AB, Ruiz M, Daniels E, Maree L, Pariyar S, Busch W, Pereira TD. Fast and Efficient Root Phenotyping via Pose Estimation. PLANT PHENOMICS (WASHINGTON, D.C.) 2024; 6:0175. [PMID: 38629082 PMCID: PMC11020144 DOI: 10.34133/plantphenomics.0175] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 11/20/2023] [Accepted: 03/20/2024] [Indexed: 04/19/2024]
Abstract
Image segmentation is commonly used to estimate the location and shape of plants and their external structures. Segmentation masks are then used to localize landmarks of interest and compute other geometric features that correspond to the plant's phenotype. Despite its prevalence, segmentation-based approaches are laborious (requiring extensive annotation to train) and error-prone (derived geometric features are sensitive to instance mask integrity). Here, we present a segmentation-free approach that leverages deep learning-based landmark detection and grouping, also known as pose estimation. We use a tool originally developed for animal motion capture called SLEAP (Social LEAP Estimates Animal Poses) to automate the detection of distinct morphological landmarks on plant roots. Using a gel cylinder imaging system across multiple species, we show that our approach can reliably and efficiently recover root system topology at high accuracy, few annotated samples, and faster speed than segmentation-based approaches. In order to make use of this landmark-based representation for root phenotyping, we developed a Python library (sleap-roots) for trait extraction directly comparable to existing segmentation-based analysis software. We show that pose-derived root traits are highly accurate and can be used for common downstream tasks including genotype classification and unsupervised trait mapping. Altogether, this work establishes the validity and advantages of pose estimation-based plant phenotyping. To facilitate adoption of this easy-to-use tool and to encourage further development, we make sleap-roots, all training data, models, and trait extraction code available at: https://github.com/talmolab/sleap-roots and https://osf.io/k7j9g/.
Collapse
Affiliation(s)
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Wolfgang Busch
- Salk Institute for Biological Studies, La Jolla, CA 92037, USA
| | | |
Collapse
|
2
|
Wang H, Moghe GD, Kovaleski AP, Keller M, Martinson TE, Wright AH, Franklin JL, Hébert-Haché A, Provost C, Reinke M, Atucha A, North MG, Russo JP, Helwi P, Centinari M, Londo JP. NYUS.2: an automated machine learning prediction model for the large-scale real-time simulation of grapevine freezing tolerance in North America. HORTICULTURE RESEARCH 2024; 11:uhad286. [PMID: 38487294 PMCID: PMC10939402 DOI: 10.1093/hr/uhad286] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 08/28/2023] [Accepted: 12/17/2023] [Indexed: 03/17/2024]
Abstract
Accurate and real-time monitoring of grapevine freezing tolerance is crucial for the sustainability of the grape industry in cool climate viticultural regions. However, on-site data are limited due to the complexity of measurement. Current prediction models underperform under diverse climate conditions, which limits the large-scale deployment of these methods. We combined grapevine freezing tolerance data from multiple regions in North America and generated a predictive model based on hourly temperature-derived features and cultivar features using AutoGluon, an automated machine learning engine. Feature importance was quantified by AutoGluon and SHAP (SHapley Additive exPlanations) value. The final model was evaluated and compared with previous models for its performance under different climate conditions. The final model achieved an overall 1.36°C root-mean-square error during model testing and outperformed two previous models using three test cultivars at all testing regions. Two feature importance quantification methods identified five shared essential features. Detailed analysis of the features indicates that the model has adequately extracted some biological mechanisms during training. The final model, named NYUS.2, was deployed along with two previous models as an R shiny-based application in the 2022-23 dormancy season, enabling large-scale and real-time simulation of grapevine freezing tolerance in North America for the first time.
Collapse
Affiliation(s)
- Hongrui Wang
- School of Integrative Plant Science, Horticulture Section, Cornell AgriTech, Cornell University, Geneva, NY 14456, USA
| | - Gaurav D Moghe
- School of Integrative Plant Science, Plant Biology Section, Cornell University, Ithaca, NY 14850, USA
| | - Al P Kovaleski
- Plant and Agroecosystem Sciences Department, University of Wisconsin–Madison, Madison, WI 53706, USA
| | - Markus Keller
- Department of Viticulture and Enology, Irrigated Agriculture Research and Extension Center, Washington State University, Prosser, WA 99350, USA
| | - Timothy E Martinson
- School of Integrative Plant Science, Horticulture Section, Cornell AgriTech, Cornell University, Geneva, NY 14456, USA
| | - A Harrison Wright
- Kentville Research and Development Centre, Agriculture and Agri-Food Canada, Kentville, Nova Scotia, B4N 1J5, Canada
| | - Jeffrey L Franklin
- Kentville Research and Development Centre, Agriculture and Agri-Food Canada, Kentville, Nova Scotia, B4N 1J5, Canada
| | | | - Caroline Provost
- Centre de Recherche Agroalimentaire de Mirabel, Mirabel, Québec, J7N 2X8, Canada
| | - Michael Reinke
- Southwest Michigan Research and Extension Center, Michigan State University, Benton Harbor, MI 49022, USA
| | - Amaya Atucha
- Plant and Agroecosystem Sciences Department, University of Wisconsin–Madison, Madison, WI 53706, USA
| | - Michael G North
- Plant and Agroecosystem Sciences Department, University of Wisconsin–Madison, Madison, WI 53706, USA
| | - Jennifer P Russo
- School of Integrative Plant Science, Horticulture Section, Cornell AgriTech, Cornell University, Geneva, NY 14456, USA
| | - Pierre Helwi
- Martell & Co., 7 place Edouard Martell, Cognac 16100, France
| | - Michela Centinari
- Department of Plant Science, The Pennsylvania State University, University Park, PA 16802, USA
| | - Jason P Londo
- School of Integrative Plant Science, Horticulture Section, Cornell AgriTech, Cornell University, Geneva, NY 14456, USA
| |
Collapse
|
3
|
Berrigan EM, Wang L, Carrillo H, Echegoyen K, Kappes M, Torres J, Ai-Perreira A, McCoy E, Shane E, Copeland CD, Ragel L, Georgousakis C, Lee S, Reynolds D, Talgo A, Gonzalez J, Zhang L, Rajurkar AB, Ruiz M, Daniels E, Maree L, Pariyar S, Busch W, Pereira TD. Fast and efficient root phenotyping via pose estimation. BIORXIV : THE PREPRINT SERVER FOR BIOLOGY 2023:2023.11.20.567949. [PMID: 38045278 PMCID: PMC10690188 DOI: 10.1101/2023.11.20.567949] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 12/05/2023]
Abstract
Image segmentation is commonly used to estimate the location and shape of plants and their external structures. Segmentation masks are then used to localize landmarks of interest and compute other geometric features that correspond to the plant's phenotype. Despite its prevalence, segmentation-based approaches are laborious (requiring extensive annotation to train), and error-prone (derived geometric features are sensitive to instance mask integrity). Here we present a segmentation-free approach which leverages deep learning-based landmark detection and grouping, also known as pose estimation. We use a tool originally developed for animal motion capture called SLEAP (Social LEAP Estimates Animal Poses) to automate the detection of distinct morphological landmarks on plant roots. Using a gel cylinder imaging system across multiple species, we show that our approach can reliably and efficiently recover root system topology at high accuracy, few annotated samples, and faster speed than segmentation-based approaches. In order to make use of this landmark-based representation for root phenotyping, we developed a Python library (sleap-roots) for trait extraction directly comparable to existing segmentation-based analysis software. We show that landmark-derived root traits are highly accurate and can be used for common downstream tasks including genotype classification and unsupervised trait mapping. Altogether, this work establishes the validity and advantages of pose estimation-based plant phenotyping. To facilitate adoption of this easy-to-use tool and to encourage further development, we make sleap-roots, all training data, models, and trait extraction code available at: https://github.com/talmolab/sleap-roots and https://osf.io/k7j9g/.
Collapse
Affiliation(s)
| | - Lin Wang
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Hannah Carrillo
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Kimberly Echegoyen
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Mikayla Kappes
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Jorge Torres
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Angel Ai-Perreira
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Erica McCoy
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Emily Shane
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Charles D. Copeland
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Lauren Ragel
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | | | - Sanghwa Lee
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Dawn Reynolds
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Avery Talgo
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Juan Gonzalez
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Ling Zhang
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Ashish B. Rajurkar
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Michel Ruiz
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Erin Daniels
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Liezl Maree
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Shree Pariyar
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Wolfgang Busch
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| | - Talmo D. Pereira
- Salk Institute for Biological Studies, La Jolla, CA 92037 United States of America
| |
Collapse
|