Limits...
Characterization of mature maize (Zea mays L.) root system architecture and complexity in a diverse set of Ex-PVP inbreds and hybrids.

Hauck AL, Novais J, Grift TE, Bohn MO - Springerplus (2015)

Bottom Line: For each trait, per se line effects were highly significant and the most important contributor to trait performance.The interaction between the environment and the additive line effect was also significant for all traits.Inbreds with contrasting effects on complexity and architecture traits were observed, suggesting that root complexity and architecture traits are inherited independently.

View Article: PubMed Central - PubMed

Affiliation: Department of Crop Sciences, University of Illinois, 1102 S. Goodwin Ave., Urbana, IL 61801 USA.

ABSTRACT
The mature root system is a vital plant organ, which is critical to plant performance. Commercial maize (Zea mays L.) breeding has resulted in a steady increase in plant performance over time, along with noticeable changes in above ground vegetative traits, but the corresponding changes in the root system are not presently known. In this study, roughly 2500 core root systems from field trials of a set of 10 diverse elite inbreds formerly protected by Plant Variety Protection plus B73 and Mo17 and the 66 diallel intercrosses among them were evaluated for root traits using high throughput image-based phenotyping. Overall root architecture was modeled by root angle (RA) and stem diameter (SD), while root complexity, the amount of root branching, was quantified using fractal analysis to obtain values for fractal dimension (FD) and fractal abundance (FA). For each trait, per se line effects were highly significant and the most important contributor to trait performance. Mid-parent heterosis and specific combining ability was also highly significant for FD, FA, and RA, while none of the traits showed significant general combining ability. The interaction between the environment and the additive line effect was also significant for all traits. Within the inbred and hybrid generations, FD and FA were highly correlated (rp ≥ 0.74), SD was moderately correlated to FD and FA (0.69 ≥ rp ≥ 0.48), while the correlation between RA and other traits was low (0.13 ≥ rp ≥ -0.40). Inbreds with contrasting effects on complexity and architecture traits were observed, suggesting that root complexity and architecture traits are inherited independently. A more comprehensive understanding of the maize root system and the way it interacts with the environment will be useful for defining adaptation to nutrient acquisition and tolerance to stress from drought and high plant densities, critical factors in the yield gains of modern hybrids.

No MeSH data available.


Vertical image of a hybrid maize root system. b Color image in (a) was processed into a binary black and white image. c The image in (b) was used for calculating the box dimension. The log–log graph relates box side length r to the number of boxes N needed to cover the root shown in (b). The slope of the “space filling box count” line is the box dimension of the root in image (b). d This plot shows the local box dimensions for image (b). This graph indicates that N and r are related by the power law  within the scale range of 1–256 pixels. All images and graphs were produced using the public Matlab program “boxcount.m”.
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4537857&req=5

Fig4: Vertical image of a hybrid maize root system. b Color image in (a) was processed into a binary black and white image. c The image in (b) was used for calculating the box dimension. The log–log graph relates box side length r to the number of boxes N needed to cover the root shown in (b). The slope of the “space filling box count” line is the box dimension of the root in image (b). d This plot shows the local box dimensions for image (b). This graph indicates that N and r are related by the power law within the scale range of 1–256 pixels. All images and graphs were produced using the public Matlab program “boxcount.m”.

Mentions: While this study did not explore the question of whether maize roots can be conclusively classified as fractal objects, our assumption that roots are fractal-like helped us to overcome limitations caused by the analysis of partial root systems. If roots are indeed fractals, loss of fine structure during the washing processes will not alter the fractal dimension estimate of a root system. In addition, the use of fractals facilitates extrapolation of root complexity to the unobserved portions of the root system. The box counting method is the standard approach applied to determine the fractal dimension of root systems. Box counting algorithms tally the number of boxes N of a given side length s needed to cover an object in the image. If this object is a fractal, N and s are related by the power law , where Dim is the “box dimension” of the object and plotting versus results in points on a straight line with Dim being the slope of this line (Fig. 4c). With experimental data, Dim is commonly estimated by fitting the linear model using the least-square approach to the data (Clauset et al. 2009). However, least-square fits do not provide information whether the data was indeed sampled from a power-law distribution. Clauset et al. (2009) developed a goodness-of-fit test for large data sets based on the Kolmogorov–Smirnov statistic and likelihood ratios. How these goodness-of-fit tests function with small data sets, like in our case, where each image provides only a set of ten data points using the box counting method to determine fractal dimensions, is yet unknown.Fig. 4


Characterization of mature maize (Zea mays L.) root system architecture and complexity in a diverse set of Ex-PVP inbreds and hybrids.

Hauck AL, Novais J, Grift TE, Bohn MO - Springerplus (2015)

Vertical image of a hybrid maize root system. b Color image in (a) was processed into a binary black and white image. c The image in (b) was used for calculating the box dimension. The log–log graph relates box side length r to the number of boxes N needed to cover the root shown in (b). The slope of the “space filling box count” line is the box dimension of the root in image (b). d This plot shows the local box dimensions for image (b). This graph indicates that N and r are related by the power law  within the scale range of 1–256 pixels. All images and graphs were produced using the public Matlab program “boxcount.m”.
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4537857&req=5

Fig4: Vertical image of a hybrid maize root system. b Color image in (a) was processed into a binary black and white image. c The image in (b) was used for calculating the box dimension. The log–log graph relates box side length r to the number of boxes N needed to cover the root shown in (b). The slope of the “space filling box count” line is the box dimension of the root in image (b). d This plot shows the local box dimensions for image (b). This graph indicates that N and r are related by the power law within the scale range of 1–256 pixels. All images and graphs were produced using the public Matlab program “boxcount.m”.
Mentions: While this study did not explore the question of whether maize roots can be conclusively classified as fractal objects, our assumption that roots are fractal-like helped us to overcome limitations caused by the analysis of partial root systems. If roots are indeed fractals, loss of fine structure during the washing processes will not alter the fractal dimension estimate of a root system. In addition, the use of fractals facilitates extrapolation of root complexity to the unobserved portions of the root system. The box counting method is the standard approach applied to determine the fractal dimension of root systems. Box counting algorithms tally the number of boxes N of a given side length s needed to cover an object in the image. If this object is a fractal, N and s are related by the power law , where Dim is the “box dimension” of the object and plotting versus results in points on a straight line with Dim being the slope of this line (Fig. 4c). With experimental data, Dim is commonly estimated by fitting the linear model using the least-square approach to the data (Clauset et al. 2009). However, least-square fits do not provide information whether the data was indeed sampled from a power-law distribution. Clauset et al. (2009) developed a goodness-of-fit test for large data sets based on the Kolmogorov–Smirnov statistic and likelihood ratios. How these goodness-of-fit tests function with small data sets, like in our case, where each image provides only a set of ten data points using the box counting method to determine fractal dimensions, is yet unknown.Fig. 4

Bottom Line: For each trait, per se line effects were highly significant and the most important contributor to trait performance.The interaction between the environment and the additive line effect was also significant for all traits.Inbreds with contrasting effects on complexity and architecture traits were observed, suggesting that root complexity and architecture traits are inherited independently.

View Article: PubMed Central - PubMed

Affiliation: Department of Crop Sciences, University of Illinois, 1102 S. Goodwin Ave., Urbana, IL 61801 USA.

ABSTRACT
The mature root system is a vital plant organ, which is critical to plant performance. Commercial maize (Zea mays L.) breeding has resulted in a steady increase in plant performance over time, along with noticeable changes in above ground vegetative traits, but the corresponding changes in the root system are not presently known. In this study, roughly 2500 core root systems from field trials of a set of 10 diverse elite inbreds formerly protected by Plant Variety Protection plus B73 and Mo17 and the 66 diallel intercrosses among them were evaluated for root traits using high throughput image-based phenotyping. Overall root architecture was modeled by root angle (RA) and stem diameter (SD), while root complexity, the amount of root branching, was quantified using fractal analysis to obtain values for fractal dimension (FD) and fractal abundance (FA). For each trait, per se line effects were highly significant and the most important contributor to trait performance. Mid-parent heterosis and specific combining ability was also highly significant for FD, FA, and RA, while none of the traits showed significant general combining ability. The interaction between the environment and the additive line effect was also significant for all traits. Within the inbred and hybrid generations, FD and FA were highly correlated (rp ≥ 0.74), SD was moderately correlated to FD and FA (0.69 ≥ rp ≥ 0.48), while the correlation between RA and other traits was low (0.13 ≥ rp ≥ -0.40). Inbreds with contrasting effects on complexity and architecture traits were observed, suggesting that root complexity and architecture traits are inherited independently. A more comprehensive understanding of the maize root system and the way it interacts with the environment will be useful for defining adaptation to nutrient acquisition and tolerance to stress from drought and high plant densities, critical factors in the yield gains of modern hybrids.

No MeSH data available.