[en] Background: Genetic analyses of plant root systems require large datasets of extracted architectural traits. To quantify such traits from images of root systems, researchers often have to choose between automated tools (that are prone to error and extract only a limited number of architectural traits) or semi-automated ones (that are highly time consuming).
Findings: We trained a Random Forest algorithm to infer architectural traits from automatically extracted image descriptors. The training was performed on a subset of the dataset, then applied to its entirety. This strategy allowed us to (i) decrease the image analysis time by 73% and (ii) extract meaningful architectural traits based on image descriptors. We also show that these traits are sufficient to identify Quantitative Trait Loci that had previously been discovered using a semi-automated method.
Conclusions: We have shown that combining semi-automated image analysis with machine learning algorithms has the power to increase the throughput of large scale root studies. We expect that such an approach will enable the quantification of more complex root systems for genetic studies. We also believe that our approach could be extended to other areas of plant phenotyping.
Herder GD, Van Isterdael G, Beeckman T et al. The roots of a new green revolution. Trends Plant Sci 2010;15:600-7.
Lynch JP. Turner review no. 14. Roots of the second green revolution. Aust J Bot 2007;55:493-512.
Lobet G, Koevoets IT, Noll M et al. Using a structural root system model to evaluate and improve the accuracy of root image analysis pipelines. Front Plant Sci 2017; doi: 10.3389/fpls.2017.00447.
Hund A, Trachsel S, Stamp P. Growth of axile and lateral roots of maize: I. Development of a phenotying platform. Plant Soil 2009;325:335-49.
Atkinson JA, Wingen LU, Griffiths M et al. Phenotyping pipeline reveals major seedling root growth QTL in hexaploid wheat. J Exp Bot Soc Experiment Biol 2015;66:2283-92.
Gioia T, Galinski A, Lenz H et al. GrowScreen-PaGe, a noninvasive, high-throughput phenotyping system based on germination paper to quantify crop phenotypic diversity and plasticity of root traits under varying nutrient supply. Funct Plant Biol 2016;44(1):76-93.
Pound MP, French AP, Atkinson JA et al. RootNav: navigating images of complex root architectures. Plant Physiol 2013;162:1802-14.
Lobet G, Pag`es L, Draye X. A novel image-analysis toolbox enabling quantitative analysis of root system architecture. Plant Physiol Am Soc Plant Biol 2011;157:29-39.
Leitner D, Felderer B, Vontobel P et al. Recovering root system traits using image analysis exemplified by twodimensional neutron radiography images of lupine. Plant Physiol 2014;164:24-35.
Cai J, Zeng Z, Connor JN et al. RootGraph: a graphic optimization tool for automated image analysis of plant roots. EXBOTJ 2015;66:6551-62.
Minervini M, Scharr H, Tsaftaris SA. Image analysis: the new bottleneck in plant phenotyping [applications corner]. IEEE Signal Process Mag 2015;32:126-31.
Ma C, Zhang HH, Wang X. Machine learning for big data analytics in plants. Trends Plant Sci 2014;19:798-808.
Ali I, Greifeneder F, Stamenkovic J et al. Review of machine learning approaches for biomass and soil moisture retrievals from remote sensing data. Remote Sensing 2015;7:16398-421.
Babatunde OH, Armstrong L, Leng J et al. A computerbased vision systems for automatic identification of plant species using kNN and genetic PCA. J Ag Inform 2015; doi:10.17700/jai.2015.6.1.152.
Minervini M, Abdelsamea MM, Tsaftaris SA. Image-based plant phenotyping with incremental learning and active contours. Ecol Inform 2014;23:35-48.
Singh A, Ganapathysubramanian B, Singh AK et al. Machine learning for high-throughput stress phenotyping in plants. Trends Plant Sci 2015;21:1-15.
Weiss U, Biber P, Laible S et al. Plant species classification using a 3D LIDAR sensor and machine learning. In: Ninth International Conference on Machine Learning and Applications, Washington, DC, 2010, pp. 339-45. New York, NY: IEEE, 2011.
Pound MP, Atkinson JA, Burgess AJ et al. Deep Machine Learning provides state-of-the-art performance in imagebased plant phenotyping. Gigascience 2017:6(9):1-7. doi: 10.1093/gigascience/gix083.
Broman KW, Wu H, Sen S et al. R/qtl: QTL mapping in experimental crosses. Bioinformatics 2003;19:889-90.
PRIMAL. Pipeline of root image analysis using MAchine learning. https://plantmodelling.github.io/primal/. Accessed 10 August 2017.
Guillaume Lobet. PRIMAL: Pipeline of root image analysis using machine learning. protocols.io 2017. http://dx.doi.org/10.17504/protocols.io.h7bb9in.
Atkinson JA, Lobet G, Noll M et al. Supporting data for "Combining semi-automated image analysis techniques with machine learning algorithms to accelerate large-scale genetic studies." GigaScience Database 2017. http://dx.doi.org/10.5524/100346.
