PosterPDF Available

Abstract

- Improving nutrient and water uptake in resource limited soils is a major challenge for agricultural research. - Root hairs are specialized epidermal cells that are important for nutrient uptake from the soil by increasing the root-surface area. - Digital microscopy can record root hairs as images, but has never been used for automated analysis. - We present an algorithm to automatically quantify the geometric complexity and spatial arrangement of root hairs.
Four treatments of maize: (1) under full strength of Hoagland's
solution (no stress), (2) under reduced Potassium, (3) under reduced
Nitrogen and (4) under reduced Phosphorus.
Three replicates per treatment and three images per primary root
were taken.
www.computational-plant-science.org
Automatic detection and quantification of
individual root hairs in complex arrangements
Peter Pietrzyk1, Chartinun Chutoe2, Patompong Saengwilai2, Alexander Bucksch1,3,4
1Dept. of Plant Biology, UGA; 2Dept. of Biology, Faculty of Science, Mahidol University; 3Warnell School of Forestry and Natural Resources, UGA; 4Inst. of Bioinformatics, UGA
Quantifying variation in root hair traits of individual plants
Improvement of our algorithm to distinguish intersecting root hairs
or clusters of 5 or more hairs.
Improvement of the experimental setup to reduce challenging cases
and achieve better manual measurements as baseline in maize.
Computing root hair density, surface area, length and orientation
provides ways to distinguish genotypes in full diversity panels.
Integration of our algorithms into DIRT (dirt.cyverse.org)
Extending our study to root hairs of common bean is planned.
Peter Pietrzyk
PhD student at the University of Georgia, Dept. of Plant Biology, Bucksch Lab
Peters phenotyping interest is to develop imaging algorithms that characterize
the variability across multiple architectural scales in various crop roots.He has a
background in Aerospace Engineering (B.Sc.) and Geomatics (M.Sc.).
PeterJacek.Pietrzyk@uga.edu @PeterPietrzyk
Reduced Phosphorus
Reduced Phosphorus
We automatically measured the 95th percentile of root hair length in
rice and maize under four different treatments to simulate manual
measurement protocols.
We achieved a high correlation between automatic measurements
and manual measurements for rice (Spearman’s correlation=0.83;
R-squared=0.61).
We could not validate the automated measurements of maize
(Spearman’s correlation=0.10; R-squared=0.02).
Measured traits did not vary significantly between the baseline
treatment and the stress induced treatments in rice and maize.
The results suggest that our algorithm distinguishes the set of test
images with short root hairs as in rice and potentially enables to
identify genetic control of root hair traits.
Introduction
Computational Methods
Rice Experiment
Outlook
Conclusions
Maize Experiment
Improving nutrient and water uptake in resource limited soils is a
major challenge for agricultural research.
Root hairs are specialized epidermal cells that are important for
nutrient uptake from the soil by increasing the root-surface area.
Digital microscopy can record root hairs as images, but has never
been used for automated analysis.
We present an algorithm to automatically quantify the geometric
complexity and spatial arrangement of root hairs.
We us machine learning to classify root hairs, the primary root from
which root hairs emerge and the image’s background (steps 1-2).
Eight training images were used to generate statistics to
automatically classify all root hair images taken with the same
imaging protocol.
At classified root hairs the medial axis transform is used to extract
the skeleton and diameter of root hairs (step 3).
Ambiguities at crossing root hairs are resolved by locally selecting
the combination resulting in the smallest curvature between all
intersecting components (step 4).
We developed first algorithms to measure length, density and
diameter of individual root hairs.
Four treatments of rice: (1) under full strength of Hoagland's
solution (no stress), (2) under reduced Potassium, (3) under reduced
Nitrogen and (4) under reduced Phosphorus.
Three replicates per treatment and three images per primary root
were taken.
Reduced Nitrogen
Challenges
Reduced Potassium
Wrong focal plane and lack of
contrast reduce the accuracy of
the classification.
Additionally, fungi, stain and dirt
on root hairs can reduce the
accuracy of the classification.
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.