Critical Study of Nutrient Sensing and Root System Architecture (Arabidopsis Thaliana)
In 2004, the United Nations published a report (United Nations, 2004) predicting that the world’s population would steadily increase, year on year, until at least 2050. Between 2004 and 2010 the world’s population increased by 5 million people (PRB, 2004 and PRB 2010), demonstrating the validity of this prediction. Last year, the United Nations (United Nations, 2010) published another report estimating that over 900 million people were still suffering from malnourishment, highlighting an insufficiency within the volume and distribution of food currently being produced. As a consequence, it has recently been suggested that as the population grows the number of people facing food insecurity will also increase (Den Herder et al, 2010), unless the pressures affecting food production can be alleviated.
To increase the volume of food being produced, scientists have advised that at least one of the following objectives has to be achieved (Godfray et al, 2010):
Improve the quantity and quality of crop harvesting
Identify or create more arable land (land that can be used for growing crops)
By increasing the quantity of the food we produce we should theoretically be able to feed the growing population. In recent years the use of fertilizers and pesticides have been used to enhance the quality and quantity of the crops however use of this cannot be increased due to the risks to public health (Den Herder et al, 2010), and so other options must be explored. Scientists are investigating methods with which to promote a plant’s intake of nutrients and to increase their tolerance against hostile environmental conditions such as drought. Another area scientists have been exploring is that of how to convert land that would be considered unsuitable for plant growth (Den Herder et al, 2010). By finding more land on which to cultivate crops, we could generate more produce using our existing techniques.
There is extensive research being conducted in both of these areas however this project will only be focussing on the work being contributed by the Plant Science Group. This Group are part of the Institute of Molecular, Cell and Systems Biology within the College of Medical, Veterinary and Life Sciences at the University of Glasgow. The Plant Science Group consists of over 50 active scientists, of which Dr Anna Amtmann is currently a Research Group Leader, and their work covers a range of topics from plant nutrition to plants effect on human metabolism. Within the Plant Science Group, Dr Amtmann’s Research Group is conducting research into the plant Arabidopsis Thaliana, with the aim of “understanding the molecular mechanisms involved within (Amtmann, 2011)”:
Nutrient usage efficiency
Salt and drought tolerance
Interaction between abiotic and biotic stress
To achieve this understanding experimental research is being conducted in a number of areas by her team, and this project aims to support the work within Nutrient Sensing and Root System Architecture.
This research involves growing Arabidopsis Thaliana in controlled environmental conditions and analyzing the effect on the plants by monitoring their individual root structures. By comparing root system architectures that have developed in different environmental conditions, specific traits can identified which can lead to the understanding of how plants intake different nutrients and respond to specific stresses.
Currently, the team grows hundreds of samples Arabidopsis Thaliana and records information about each of their root system architectures. This results in large volumes of data that require analysis and documentation. Currently the team create graphs to visualise and analyse this data but they have recently advised of the unsuitability of this approach. This is predominantly due to the limited visualization options of graphs when considering root parameters and the fact that graphs can extremely time-consuming to construct.
Therefore, this project aims to develop a system that will allow Dr Amtmann’s Research Group to visualise root system architectures of Arabidopsis Thaliana by using techniques that will manipulate the root structure architectural data into interesting and understandable representations. The following chapters of this proposal aim to document the proposed work that will be undertaken in a master project that will achieve this aim.
2. Problem Statement
Due to the biological nature of this project, this section will define a number of key concepts required to understand the work undertaken by Dr Amtmann and her colleagues, and the issues they face, before discussing the intentions of this project.
Root System Architecture
Dr Amtmann’s team assesses the root system architectures of Arabidopsis Thaliana to understand how environmental conditions affect its development. The root system of a plant is responsible for the intake of water and nutrition (Malamy, 2005), among other essential functions, and so plays an important role within the development of any plant. The root system normally consists of a main root with a number of lateral branches, which contain a number of root hairs that are responsible for the uptake of nutrients and water (Wikipedia, 2011). The way in which these underlying roots expand and branch is considered to be the root system architecture (RSA) (Figure 1). Depending on the environment in which a plant resides, changes occur to the RSA that can have a significant effect on the water or nutrient uptake of the plant(Armengaud et al, 2009). Therefore by understanding the environmental conditions that promote the growth of a successful RSA we can ensure the development of a healthy plant.
Figure 1 – Root Structure Architecture (RSA) taken from “Root-System Development and Water-Extraction Model Considering Hydrotropism” by D. Tsutsumi et al.
Like many other research projects, Amtmann’s group uses the plant Arabidopsis Thaliana within their investigations. Arabidopsis Thaliana is considered to be a model species within plant science and biology as its genome was sequenced in 2000, meaning that its complete DNA sequence was determined (Wikipedia, 2011). As the genome is relatively small it is easier to work with and analyse. It is also closely related to a large number of other plants such as (Somerville and Koorneef, 2002) meaning that any finding could be widely applicable. It can also be easily grown in laboratory conditions due to its small size and straightforward growth conditions.
As a result of the Arabidopsis Thaliana sequenced genome, scientists can determine which genetic combinations are responsible for certain characteristics of the plant, such as the direction of root growth or number of lateral branches. Figure 2 highlights some variation in roots of Arabidopsis Thaliana.
Figure 2 – Different Root Structure Architecture for Arabidopsis taken from http://www.unil.ch/dbmv/page36242_en.html
Currently Dr Amtmann’s group grows multiple instances of Arabidopsis Thaliana in square agar plates that allow them to take two-dimensional images of the plant roots as it grows. These images are then examined using a bespoke system called EZ-Rhizo, which was developed by Dr Amtmann’s Research Group, to determine the RSA of the plant.
EZ-Rhizo is freely available software that detects and measures the RSA of a plant quickly and accurately (Armengaud et al, 2009) when supplied with an image of a root system. It was developed by Dr Amtmann’s Research Group in conjunction with undergraduate students from l’Ecole Superieure d’Ingenieurs en Electrotechnique et Electrique in 2008. The main function of the system is to measure information about the main root and the lateral roots of a plant, and record this. All numerical values are stored in centimetres. It then uses this data to derive further parameters such as the apical zone or straightness of the root. The key parameters that are captured by EZ-Rhizo in relation to this project are as follows:
Apical Zone, which is the length of the root between the top of the main root and the oldest lateral root
Length, which is the total length of the root from the origin of the main root to the tip of it. (Note that as roots do not grow straight this value may be very large but the root may not have grown that far towards the ground)
Vector Length, this is similar to the length however it calculate the shortest possible length of the root starting from the origin of the main root to the tip of it.
Straightness, which is the Vector Length, divided by the Length.
Angle, which is the degree between the Vector of the root and complete verticality.
A full listing of all the parameter capture by EZ-Rhizo can located in the in that’s publication
The system was developed for the Microsoft Windows operating system and stores data in a MySQL database. Queries can then be submitted to the database, which will retrieve all of the stored data for a specified root and output it into a Comma Separated Values (CSV) file (Armengaud et al, 2009).
As previously described, the main method used by Dr Amtmann’s Group to analyse and compare the plant root data, recorded by EZ-Rhizo, is to gather the required CSV files and generate the appropriate graphs in Microsoft Excel. The current methods used by this group can be extremely time consuming and could be subject to error if mislabelling on a graph occurs or if a value within a CSV file is misread. As the team work with large quantities of data there is a high probability of this occurring. Also, the format of the results could also inhibit certain comparisons, as the visualization of the roots is limited to the graph format. Therefore the aim of this project is to design and implement a new system that will enable Dr Amtmann’s Research Group to overcome these difficulties.
The proposed system will allow them to analyse and visualise their root data, collected from EZ-Rhizo, in a more flexible and reliable manner. The basic functionality of this system will be to create a realistic visualization of a root based on the supplied parameters. This will provide the foundation for developing alternative visualizations but it will also allow the researchers to understand the data that EZ-Rhizo is recording. As the system will re-create the root system architecture based on only the values provided by EZ-Rhizo, the re-creations will highlight which characterises of the root can be displayed. Therefore, once the visualizations have been created it may show that more root data is required to be captured by EZ-Rhizo in order to for the root system architecture to be properly documented. This verification could not be achieved from a graph of the values or from viewing the original root structure image that is supplied to EZ-Rhizo.
Once the basic functionality of this system is operational it can be enhanced, by the use of prototypes and experimentation, to find suitable techniques that allow for a meaningful representation of more than one root. For example, the average value of each parameter could be determined for a set of roots and the system could draw the average root. Alternatively given the parameters for a set of roots, the system may be able to estimate a root visualization that had all of those parameters altered by a set value and so produce a theoretical root. Estimations of this nature would not be possible using the group’s current graphical techniques.
In summary, the objective of this project is to implement a system that will create interesting and realistic visualizations of plants root system architectures that will be used by Dr Anna Amtmann and her Research Group in their experiments.
3. Background Survey
One of the functional requirements that was supplied by Dr Anna Amtmann’s Research Group during the initial discussions of this project was that they require this system to be multi-platform, unlike EZ-Rhizo, which was specifically designed for Microsoft Windows Operating Systems.
As a result of this requirement and the author’s previous development experience, the Java development language has been selected for this project. As Java is a cross-platform language that be run anywhere, as long as the Java Virtual Machine (JVM) is present, it meet the requirement set by the Research Group. Also, as a main priority of this project is to develop and experiment with the greatest number of prototypes, it is advisable to reduce the learn curve that would be required in understanding a new language and so as Java is the developers strongest language it has been selected.
The Java 2D is a freely available drawing framework that allows for the construction of two-dimensional images (Wikipedia, 2011) and is part of the core Java Standard Edition Application Programming Interfaces (APIs) developed by Sun Microsystems, and now maintained by Oracle (Oracle, 2011).The key strengths of this API are that it supports the drawing of basic shapes such as rectangles and ellipses but also basic lines, which is required for this project. Also, it will export any images created into well-known formats such as JPEG or GIF (Sun Developer Network, 2011). As this is a JAVA API there is also a great deal of implementation support provided.
Open Graphics Language (OpenGL) is a mutli-platform API for creating two-dimensional or three-dimensional graphics (OpenGL, 2011) that is now considered to be an industry standard for graphical development. Currently OpenGL can be used with a number of development languages including Java (JOGL), Python, C++ and C. As this project will be constructed in Java, JOGL would be the appropriate selection and since the release of Java Standard Edtion 6 in December 2006, JOGL and Java 2D can be used concurrently. However, one limitation of JOGL is that it will add complexity to a if the developers has no previous experience with this API as it currently has 250 different function calls for drawing one image (Wikipedia, 2011).
One of the key elements of the root visualization is to get it to be as realistic as possible. Now it is likely that the root will be drawn out of straight lines how it will be essential that
4. Proposed Approach
The suggested approach for this project will involve designing and implementing a basic system that contains all of the functionality required to visualize one root based on the information provided from EZ-Rhizo. Upon successfully implementation of this initial system, a number of prototypes will be created around techniques, such as Alpha Blending, that will manipulate this functionality to draw clusters of roots or alter the initial visualization to provide a different perspective for the user. Any prototype functionality that is considered to be useful, interesting or successful will then be fully integrated into the main system as additional functionality.
It is likely that the new system will send requests directly to the database to retrieve this information rather than obtaining this from CSV files. However this functionality will be discussed with Dr Amtmann’s group during the design phase of the basic system alongside all other functional and non-functional requirements
This system will only work with data from EZ-Rhizo at this time and it is assumed that the parameters will be from Arabidopsis Thaliana
JAVA – it needs to be multi-platform unlike Ez-rhio and JDBC
Currently at least two general prototypes have been envisioned with the following functionality
Being able to represent average values for root parameters, such as length, and display them in a readable and understandable manner
Being able to represent the lifecycle of a root based on the parameters recorded on the first day of growth to the last.
It is likely that each of these general prototypes will have sub-prototypes that will evaluate techniques that display the information to achieve the objective. After the more successful techniques have been identified they will be developed and presented to the research group to confirm if they still wish this prototype to be included.
To ensure the success of this project, the implementation of this project will follow an iterative software development methodology of design, implementation and testing. This has been selected because one of the key components of this project is to build upon a basic system. It is therefore essential that this be successfully deployed before introducing any further functionality from desired prototypes. Due to this requirement the selected methodology will be The Spiral, which supports iterations and constant reviews and refinements as shown in Figure 3. It is also vital to review each prototype after it has been implemented to ensure that the original functionality has not been compromised.
Figure 3 – Spiral Software Development Methodology taken from http://as.exeter.ac.uk/divisions/exeterit/iws/projects/ede/
As with any software development project there is a certain element of risk in undertaking the assignment, and the key risks identified within this project are as follows:
Due to biological terminology involved in the understanding of the system requirements there is a possibility that these may be interpreted wrongly. Therefore it is imperative that this be considered at every stage involving an element of design, and clarification will be sought if any ambiguity arises.
As this project is due to be completed over a 15-week period there is potential for the project to overrun at any point and so continual review of the work plan is required. The stage with the highest potential for delay would be that of developing and testing prototypes. Once the initial prototypes have been produced there is a possibility that revision will be required after discussions with Dr Amtmann’s Research Team. Where possible all prototypes will be implemented to the highest standard however at a given point the further development of the prototypes will have to be discontinued. Regular discussions with Dr Rogers and Dr Amtmann regarding this should identify a suitable stopping point and prevent the project from suffering from delays.
5. Work Plan
This chapter will detail the provisional work plan for this project and it has been estimated that work will begin on 20th June 2011. Please note that weekly meetings will be scheduled with Dr Simon Rogers during this time and there will be a high level of communication between Dr Anna Amtmann’s Research Group and myself.
Figure 4 – Gantt Chart Showing Estimated Project Timescale
The following sections detail the work that will be done in each phase of the project as shown in the Gantt chart in Figure 4.
During this time I will study the areas identified in section three in more depth and review this background research. At this stage I will also investigate further techniques that could be used to develop additional prototypes that could be used to manipulate and display the root data in an interesting manner.
Deliverables: Revised Background Section
Critical Rating: Preferably
Risk Factor: Low
This stage will involve identifying and creating a detailed list of the functional and non-functional requirements of the system. During this time a requirements document containing the appropriate UML diagrams will be developed and will be approved by Dr Simon Rogers and Dr Anna Amtmann before any implementation begins.
Deliverables: A clear requirements document
Critical Rating: Important
Risk Factor: Medium
Implementation of the Basic Functionality
After the basic requirements of the system have been determined they will be implemented in an application. At this stage, the system should at a minimum be able to use the data from EZ-Rhizo and create a visualization of at least one root. A test plan will also be created at this time.
Deliverables: Implementation of Basic System and Test Plan
Critical Rating: Fundamental
Risk Factor: Medium
Testing of Basic Functionality
This stage will ensure that the basic system is operational and that all of the functional requirements have been achieved. This may involve getting feedback from the Plant Science Group depending on their availability and whether or not they wish to test the completed project.
Critical Rating: Important
Risk Factor: Low
Design and Build Prototypes
The next section will involve developing prototypes that manipulate and display the root data in various interesting manners. Each prototype will involve a technique identified previously in section three or found within the further research conducted at the beginning of the project plan.
Deliverables: A number of prototypes that modify the visualization of a basic root
Critical Rating: Fundamental
Risk Factor: Low
Experiments with Prototypes
At this time Dr Amtmann and her team will review the prototypes with a follow up discussion with Dr Rogers and myself to confirm that they capture useful information. At this time, depending on time constraints, additional prototypes may be developed based on the discussions, however this will be reviewed at the time.
Deliverables: A list of prototypes that are deemed to be visually interesting and useful
Critical Rating: Preferable
Risk Factor: Medium
Integrate desired Prototypes
Any prototypes that are deemed to be successful will then be integrated into the basic system. This will be completed in iterations, with testing being conducted after each integration to ensure that the new prototype does not break the basic system.
Critical Rating: Important
Risk Factor: High
Testing of Additional Functionality
Once every prototype has been implemented successfully, further testing will be conducted to ensure that the original functionality still meets the initial criteria and that the prototype functionality is producing valid and interesting results. At this stage a small user evaluation may be conducted to ensure that the user interface developed for the system is successful.
Critical Rating: Important
Risk Factor: Low
Complete and Submit Dissertation
Finally, a dissertation documenting the work that has been undertaken will be composed and will be approximately 60-70 pages long. It will be submitted on the 6th September 2011. During the project, work will be provisionally documented to ensure an easier transition from implementation to documentation.
Deliverables: A complete dissertation documenting the progress and outcome of this project
Critical Rating: Fundamental
Risk Factor: Low
(Armengaud et al, 2009)
Armengaud, P., Zambaux, K., Hills, A., Sulpice, R., Pattison, R. J., Blatt, M. R. and Amtmann, A. (2009), EZ-Rhizo: integrated software for the fast and accurate measurement of root system architecture. The Plant Journal, 57: pp. 945–956.
(Den Herder et al, 2010)
Den Herder, G., Van Isterdael, G., Beeckman, T. and De Smet, I. (2010), The roots of a new green revolution, Trends in Plant Science, Volume 15, Issue 11, November 2010, pp. 600-607.
(Godfray et al, 2010)
Godfray, H.C., Beddington, J.R., Crute, I.R., Haddad, L., Lawrence, D, Muir, J.F., Pretty, J. Thomas, S.M. and Toulmin, C. (2010), The challenge of feeding 9 billion people. Science Magazine, 12th February 2010: 327 (5967), pp. 812-818.
Malamy, J. E. (2005), Intrinsic and environmental response pathways that regulate root system architecture. Plant, Cell & Environment, 28: pp. 67–77.
Population Reference Bureau, www.prb.org/pdf04/04worlddatasheet_eng.pdf, accessed 20th March 2011
(Population Reference Bureau, 2010)
Population Reference Bureau, www.prb.org/Publications/Datasheets/2010/2010wpds.aspx, accessed 20th March 2011
(Somerville and Koorneef, 2002)
Somerville, C. and Koornneef, M. (2002), A fortunate choice: the history of Arabidopsis as a model plant. Nature Reviews Genetics, 3:11, pp. 883 – 889
(United Nations, 2004)
United Nations Department of Economic and Social Affairs: Population Division, http://www.un.org/esa/population/publications/longrange2/WorldPop2300final.pdf, accessed 2nd March 2011
(United Nations, 2010)
Food and Agriculture Organization of United Nations, www.fao.org/hunger/en/, accessed 2nd March 2011
Wikipedia. Root – Wikipedia, the free encyclopaedia. Retrieved March 2011 from
Wikipedia. Genome – Wikipedia, the free encyclopaedia. Retrieved March 2011 from http://en.wikipedia.org/wiki/Genome_sequence