Jonathan A Atkinson

Dr. Jonathan A Atkinson


Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
23 August 2017 in GigaScience
Verified
Journal Article
Open access
PDF file
MPMichael P. Pound
Michael P. Pound
JAJonathan A Atkinson
Jonathan A Atkinson
ATAlexandra J. Townsend
Alexandra J. Townsend
MWMichael H. Wilson
Michael H. Wilson
MGMarcus Griffiths
Marcus Griffiths
AFAndrew French
Andrew French
In plant phenotyping, it has become important to be able to measure many features on large image sets in order to aid genetic discovery. The size of the datasets, now often captured robotically, often precludes manual inspection, hence the motivation for finding a fully automated approach. Deep learning is an emerging field that promises unparalleled results on many data analysis problems. Building on artificial neural networks, deep approaches have many more hidden layers in the network, and hence have greater discriminative and predictive power. We demonstrate the use of such approaches as part of a plant phenotyping pipeline. We show the success offered by such techniques when applied to the challenging problem of image-based plant phenotyping and demonstrate state-of-the-art results (>97% accuracy) for root and shoot feature identification and localization. We use fully automated trait identification using deep learning to identify quantitative trait loci in root architecture datasets. The majority (12 out of 14) of manually identified quantitative trait loci were also discovered using our automated approach based on deep learning detection to locate plant features. We have shown deep learning–based phenotyping to have very good detection and localization accuracy in validation and testing image sets. We have shown that such features can be used to derive meaningful biological traits, which in turn can be used in quantitative trait loci discovery pipelines. This process can be completely automated. We predict a paradigm shift in image-based phenotyping bought about by such deep learning approaches, given sufficient training sets.

1 November 2019 in GigaScience
Verified
Journal Article
Open access
PDF file
Robail YasrabRobail Yasrab
Robail Yasrab
JAJonathan A Atkinson
Jonathan A Atkinson
DWDarren M. Wells
Darren M. Wells
AFAndrew French
Andrew French
TPTony P Pridmore
Tony P Pridmore
MPMichael Pound
Michael Pound
In recent years quantitative analysis of root growth has become increasingly important as a way to explore the influence of abiotic stress such as high temperature and drought on a plant's ability to take up water and nutrients. Segmentation and feature extraction of plant roots from images presents a significant computer vision challenge. Root images contain complicated structures, variations in size, background, occlusion, clutter and variation in lighting conditions. We present a new image analysis approach that provides fully automatic extraction of complex root system architectures from a range of plant species in varied imaging set-ups. Driven by modern deep-learning approaches, RootNav 2.0 replaces previously manual and semi-automatic feature extraction with an extremely deep multi-task convolutional neural network architecture. The network also locates seeds, first order and second order root tips to drive a search algorithm seeking optimal paths throughout the image, extracting accurate architectures without user interaction. We develop and train a novel deep network architecture to explicitly combine local pixel information with global scene information in order to accurately segment small root features across high-resolution images. The proposed method was evaluated on images of wheat (Triticum aestivum L.) from a seedling assay. Compared with semi-automatic analysis via the original RootNav tool, the proposed method demonstrated comparable accuracy, with a 10-fold increase in speed. The network was able to adapt to different plant species via transfer learning, offering similar accuracy when transferred to an Arabidopsis thaliana plate assay. A final instance of transfer learning, to images of Brassica napus from a hydroponic assay, still demonstrated good accuracy despite many fewer training images. We present RootNav 2.0, a new approach to root image analysis driven by a deep neural network. The tool can be adapted to new image domains with a reduced number of images, and offers substantial speed improvements over semi-automatic and manual approaches. The tool outputs root architectures in the widely accepted RSML standard, for which numerous analysis packages exist (http://rootsystemml.github.io/), as well as segmentation masks compatible with other automated measurement tools. The tool will provide researchers with the ability to analyse root systems at larget scales than ever before, at a time when large scale genomic studies have made this more important than ever.

1 October 2017
Verified
Conference Paper
Closed access
MPMichael P. Pound
Michael P. Pound
JAJonathan A Atkinson
Jonathan A Atkinson
DWDarren M. Wells
Darren M. Wells
TPTony P. Pridmore
Tony P. Pridmore
AFAndrew French
Andrew French
Plant phenotyping has continued to pose a challenge to computer vision for many years. There is a particular demand to accurately quantify images of crops, and the natural variability and structure of these plants presents unique difficulties. Recently, machine learning approaches have shown impressive results in many areas of computer vision, but these rely on large datasets that are at present not available for crops. We present a new dataset, called ACID, that provides hundreds of accurately annotated images of wheat spikes and spikelets, along with image level class annotation. We then present a deep learning approach capable of accurately localising wheat spikes and spikelets, despite the varied nature of this dataset. As well as locating features, our network offers near perfect counting accuracy for spikes (95.91%) and spikelets (99.66%). We also extend the network to perform simultaneous classification of images, demonstrating the power of multi-task deep architectures for plant phenotyping. We hope that our dataset will be useful to researchers in continued improvement of plant and crop phenotyping. With this in mind, alongside the dataset we will make all code and trained models available online.

10 October 2019 in Plant, Cell & Environment
Verified
Journal Article
Open access
PDF file
JAJonathan A Atkinson
Jonathan A Atkinson
MHMalcolm Hawkesford
Malcolm Hawkesford
WWWilliam Whalley
William Whalley
HZHu Zhou
Hu Zhou
SMSacha Mooney
Sacha Mooney
Deep rooting is critical for access to water and nutrients found in subsoil. However, damage to soil structure and the natural increase in soil strength with depth, often impedes root penetration. Evidence suggests that roots use macropores (soil cavities greater than 75 μm) to bypass strong soil layers. If roots have to exploit structures, a key trait conferring deep rooting will be the ability to locate existing pore networks; a trait called trematotropism. In this study, artificial macropores were created in repacked soil columns at bulk densities of 1.6 g cm−3 and 1.2 g cm−3, representing compact and loose soil. Near isogenic lines of wheat, Rht‐B1a and Rht‐B1c, were planted and root–macropore interactions were visualized and quantified using X‐ray computed tomography. In compact soil, 68.8% of root–macropore interactions resulted in pore colonization, compared with 12.5% in loose soil. Changes in root growth trajectory following pore interaction were also quantified, with 21.0% of roots changing direction (±3°) in loose soil compared with 76.0% in compact soil. These results indicate that colonization of macropores is an important strategy of wheat roots in compacted subsoil. Management practices to reduce subsoil compaction and encourage macropore formation could offer significant advantage in helping wheat roots penetrate deeper into subsoil.

4 October 2017 in Frontiers in Plant Science
Verified
Journal Article
Open access
JAJonathan A Atkinson
Jonathan A Atkinson
DWDarren M. Wells
Darren M. Wells
Quantification of the tissue and cellular structure of plant material is essential for the study of a variety of plant sciences applications. Currently, many methods for sectioning plant material are either low throughput or involve free-hand sectioning which requires a significant amount of practice. Here, we present an updated method to provide rapid and high-quality cross sections, primarily of root tissue but which can also be readily applied to other tissues such as leaves or stems. To increase the throughput of traditional agarose embedding and sectioning, custom designed 3D printed moulds were utilized to embed 5-15 roots in a block for sectioning in a single cut. A single fluorescent stain in combination with laser scanning confocal microscopy was used to obtain high quality images of thick sections. The provided CAD files allow production of the embedding moulds described here from a number of online 3D printing services. Although originally developed for roots, this method provides rapid, high quality cross sections of many plant tissue types, making it suitable for use in forward genetic screens for differences in specific cell structures or developmental changes. To demonstrate the utility of the technique, the two parent lines of the wheat (Triticum aestivum) Chinese Spring x Paragon doubled haploid mapping population were phenotyped for root anatomical differences. Significant differences in adventitious cross section area, stele area, xylem, phloem, metaxylem and cortical cell file count were found.

21 July 2025 in New Phytologist
Verified
Journal Article
Open access
FAFrancisco Dini Andreote
Francisco Dini Andreote
DWDarren M. Wells
Darren M. Wells
JAJonathan A Atkinson
Jonathan A Atkinson
BABrian S. Atkinson
Brian S. Atkinson
OFOmri M. Finkel
Omri M. Finkel
GCGabriel Castrillo
Gabriel Castrillo
Summary: Soils are highly heterogeneous and dynamic systems, experiencing a constant flow of plant root exudates and moisture fluctuations that affect nutrient distribution, soil physicochemical properties, and microbial metabolisms. Plant roots adjust their development within the soil matrix (growth, branching, root angle, and anatomical features) by integrating local environmental conditions with physiologically informed signals. These physiological responses and the adaptability of roots are commonly defined as plasticity. Although genetically determined, root plasticity is modulated by local fluctuations in water and nutrient availability, environmental changes, and soil properties as well as by the root‐associated microbiota. Roots and their vicinity are colonized by taxonomically and functionally diverse microbial species. Specific members within these communities can establish chemical communication with plants via root‐derived signals, thereby tapping into the plant's hormonal and developmental network, influencing root plasticity. Given that most root traits associated with plasticity have been discovered under axenic conditions, our knowledge of the multiple potential mechanisms by which members of the root‐associated microbiota modulate root plastic responses is still limited. In this review, we explore the recent progress in this field and identify specific avenues for future research toward advancing molecular and ecological understanding of microbial‐mediated root plasticity in terrestrial systems.