Impact Factor 3.677

The world's most-cited Plant Sciences journal

This article is part of the Research Topic

Phenomics

Editorial ARTICLE

Front. Plant Sci., 23 May 2018 | https://doi.org/10.3389/fpls.2018.00678

Editorial: Phenomics

Marcos Egea-Cortines1* and John H. Doonan2
  • 1Genética Molecular, Instituto de Biotecnología Vegetal, Universidad Politécnica de Cartagena, Cartagena, Spain
  • 2National Plant Phenomics Centre, Aberystwyth University, Aberystwyth, United Kingdom

Editorial on the Research Topic
Phenomics

Developments in high throughput molecular technologies, from DNA sequencing to metabolite analysis and proteomics, have opened up new and previously undreamt of vistas in biology. Previously, one often had to make a difficult choice between longitudinal and cross-sectional studies but, with these highly scalable technologies, the number of individuals that can be screened has increased so dramatically that temporal studies are possible on whole populations. The technologies tend not to be specific to a given species, allowing us to sample the entire tree of life. One consequence of this technological explosion is that measurement of phenotypic traits for large populations over developmental time and in response to environmental variable has become highly desirable, if not a necessity (Houle et al., 2010). In this context, the development of new technologies to obtain reliable phenotypic data is a pre-requisite to approaching the overall challenge. As compared to the genotype (or even the proteome), the phenotype is highly dimensional to the extent that measuring all possible phenotypic traits is not feasible. However, the concept of “phenomics” has been proposed to cover sets of technologies devised to obtain phenotypic data in an analogous way to ‘omics associated with the various molecular technologies. Phenomics therefore includes a vast array of approaches that, in most cases, include some sort of automatic sampling or non-invasive methods to obtain repeated sampling from an individual or population.

The general requirement for reproducibility is an additional driver for phenomics. While commercial phenotyping platforms can be very powerful (Virlet et al., 2017), the economic aspects of purchasing and maintenance and the lack of flexibility (in what are emerging technologies) has fostered in-house developments (Navarro et al., 2012; Lou et al., 2014). In plant biology, growth conditions (the environment) play a key role in the final phenotype of a plant and having well-defined growth parameters is not yet the rule (despite what the material and methods section of a typical peer-reviewed research paper might imply). In this special topic on Phenomics, Negi et al. addressed reproducibility of growth conditions, developing a modified hydroponic system to test for phosphate deficiency on rice root traits. The digital nature of the data is a major advantage as it allows sharing and re-use, both key to the success of the other ‘omics technologies. An open-source software tool (Seedusoon) allows management of germplasm gathering together phenotypic and genetic data for a given accession (Charavay et al.).

Many of the non-destructive phenomic approaches rely on image analysis systems to acquire and process images. While the approach may seem straightforward, quantitative extraction of interesting features, such as intensity of image pixels, geometry of pixels or textures, remains challenging, and trade-offs between the ideal and the affordable are commonplace. For example, a key decision involves the type of camera used for data capture as that can limit the band width used to measure a given trait. This will have knock-on consequences, affecting the procedures used for image analysis (Navarro et al., 2016; Perez-Sanz et al., 2017). In the current edition, several publications t address issues associated with the analysis of a variety of plants using different image acquisition devices. Standard cameras including those found in smartphones perform image acquisition with an RED-GREEN-BLUE or RGB sensor. One study utilizes RGB images to determine wheat density at early stages of development (Liu et al.). There is an increasing number of publicly available libraries that facilitate image analysis (see Perez-Sanz et al., 2017 for a review). OpenCV, a widely used image processing library, underpinned development of SeedCounter (Komyshev et al.). This free Android App for mobile phone and pads, provides seed and grain morphometry under lab and field conditions, with much of the functionality of much more expensive equipment.

Stereo-vision is a long-established technique that uses two carefully positioned RGB cameras to capture 3-D information. Growth has been monitored in four species of tree seedling using the green channel and a stereo-vision approach (Montagnoli et al.). A regression model between the level of “greenness” and the real biomass obtained by destructive measures gave R values ranging between 0.67 for Fagus sylvatica and 0.95 for Quercus ilex, again showing actual differences between plants for a given setup. The interaction between canopy structure and photosynthesis has been studied by coupling 3-D reconstruction with gas exchange analysis showing that even complex traits such as 3-D structures can be related to photosynthesis efficiency (Burgess et al.).

The non-visible wavelengths can provide additional information on physiology and function. Thermal infrared imaging devices mounted on unmanned aerial vehicles (UAV) enables high throughput analysis of Populus nigra populations for dynamic responses to drought stress (Ludovisi et al.). Combined hyperspectral and thermal imaging of lettuce reveals how these plants adapt to multiple stresses (Simko et al.). Hyperspectral imaging has high information content and can measure several parameters simultaneously when calibrated. Thus, parallel analysis of chlorophyll a, chlorophyll b, total chlorophyll, and carotenoid in rice showed high correlation with hand measurements is 0.827–0.928 at the tillering stage, illustrating great potential to screen large populations (Feng et al.).

Using a combination of five non-invasive camera-based imaging units equipped with fluorescent, RGB Visible Near Infrared (VNIR), Short Wave Infrared and three dimensional imaging, Lyu et al. determined a total of 200 quantitative traits during leaf senescence. This illustrates the enormous potential of phenomic approaches to have a comprehensive understanding of biological variation.

High-throughput screening of combinations of traits is the immediate promise of phenomics and is further exemplified by the use of near-infrared reflectance spectroscopic (NIRS) to undertake a coordinated analysis of oil, protein, carbon, and nitrogen content in Arabidopsis seeds. As a result, a set of QTLs controlling these traits, and the variance component of genotype, culture, Genetic by Environment interaction, and residual effect have been determined (Jasinski et al.).

Image-based approaches can be compromised by the quality of the signal obtained. This is an ongoing problem common to many ‘omics technologies where assessment of quality plays a key role in downstream data analysis. Directly addressing this problem (Lobos and Poblete-Echeverría) developed software to assess the quality of spectral reflectance data. As spectral reflectance data are widely used to obtain crop performance indices such as NDVI, this type of exploratory data analysis is essential for evaluating data quality.

Author Contributions

ME-C wrote draft. JD corrected it and both agreed on final version.

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

Work in the lab of ME-C is funded by MICINN BFU2017-88300-C2-1-R and Fundación Séneca 19398/PI/14. Work in the lab of JHD is funded by BBSRC grants BB/J004464/1; BB/M018407/1; BB/L009889/1; BB/CAP1730/1 and the European Union (EPPN Grant Agreement No. 284443 and EPPN2020 grant agreement No. 731013.

References

Houle, D., Govindaraju, D., and Omholt, S. (2010). Phenomics: the next challenge. Nat. Rev. Genet. 11, 855–866. doi: 10.1038/nrg2897

PubMed Abstract | CrossRef Full Text | Google Scholar

Lou, L., Liu, Y., Han, J., and Doonan, J. H. (2014). Accurate multi-view stereo 3D reconstruction for cost-effective plant phenotyping. Lect. Notes Comput. Sci. 8815, 349–356. doi: 10.1007/978-3-319-11755-3_39

CrossRef Full Text | Google Scholar

Navarro, P. J. P. J., Fernández, C., Weiss, J., and Egea-Cortines, M. (2012). Development of a configurable growth chamber with a vision system to study circadian rhythm in plants. Sensors 12, 15356–15375. doi: 10.3390/s121115356

PubMed Abstract | CrossRef Full Text | Google Scholar

Navarro, P. J., Pérez, F., Weiss, J., and Egea-Cortines, M. (2016). Machine learning and computer vision system for phenotype data acquisition and analysis in plants. Sensors 16:E641. doi: 10.3390/s16050641

PubMed Abstract | CrossRef Full Text | Google Scholar

Perez-Sanz, F., Navarro, P. J., and Egea-Cortines, M. (2017). Plant phenomics: an overview of image acquisition technologies and image data analysis algorithms. Gigascience 6, 1–18. doi: 10.1093/gigascience/gix092

PubMed Abstract | CrossRef Full Text | Google Scholar

Virlet, N., Sabermanesh, K., Sadeghi-Tehran, P., and Hawkesford, M. J. (2017). Field Scanalyzer: an automated robotic field phenotyping platform for detailed crop monitoring. Funct. Plant Biol. 44:143. doi: 10.1071/FP16163

CrossRef Full Text | Google Scholar

Keywords: phenomics, artificial vision, RGB data, RGB image analysis, multispectral imaging

Citation: Egea-Cortines M and Doonan JH (2018) Editorial: Phenomics. Front. Plant Sci. 9:678. doi: 10.3389/fpls.2018.00678

Received: 23 March 2018; Accepted: 03 May 2018;
Published: 23 May 2018.

Edited by:

Fabio Marroni, University of Udine, Italy

Reviewed by:

Ulrich Schurr, Forschungszentrum Jülich, Germany

Copyright © 2018 Egea-Cortines and Doonan. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Marcos Egea-Cortines, marcos.egea@upct.es