iSee Logo
INSTITUTE FOR SUSTAINABILITY, ENERGY, AND ENVIRONMENT
UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN
Select Page

Crops in silico project

Project Overview

icon-solaricon-natureThe Crops in silico project fits into the iSEE research themes of Climate Solutions and Secure and Sustainable Agriculture.

 

The Challenge

As the Earth’s population climbs toward 9 billion by 2050 — and the world climate continues to change, affecting temperatures, weather patterns, water supply, and even the seasons — future food security has become a grand world challenge. Accurate prediction of how food crops react to climate change will play a critical role in ensuring food security.

 

The Solution

An ability to computationally mimic the growth, development and response of crops to the environment will allow researchers to conduct many more experiments than can realistically be achieved in the field. Designing more sustainable crops to increase productivity depends on complex interactions between genetics, environment, and ecosystem. Therefore, creation of an in silico — computer simulation — platform that can link models across different biological scales, from cell to ecosystem level, has the potential to provide more accurate simulations of plant response to the environment than any single model could alone.

As a leader in plant biology, crop sciences and computer science, Illinois is uniquely positioned to head this initiative. Developments in high-performance computing, open-source version-controlled software, advanced visualization tools, and functional knowledge of plants make achieving the concept realistic. The interdisciplinary Crops in silico (Cis) team will take advantage of resources in the National Center for Supercomputing Applications (NCSA) and the Institute for Sustainability, Energy, and Environment (iSEE) — and its academic and research expertise in plant biology, crop sciences, and bioengineering — to build a user-friendly platform for plant scientists around the globe who are working on the food security challenge.

Logo_mix

plants in silico


Project News

As the team transitions to Cis 2.0, Principal Investigator Amy Marshall-Colón helps wrap up the Cis 1.0 project:

Current models predict an increasing gap between food supply and demand over the next 50 years. There is uncertainty that U.S. breeding programs will be able to develop cultivars within this time frame to mitigate the yield discrepancy — especially at a time when land area and water resources are limiting to agricultural productivity.

Thus, there is an immediate need for technology capable of predicting the fitness of various crop phenotypes in response to climate and resource availability.

The aim of the Crops in silico 1.0 project was to lay the foundation for generating virtual plant models that accurately capture whole-system dynamics in response to in silico environmental and genetic perturbations. A goal of this research project was to identify a feasible approach to link models across biological scales, specifically from genes to phenotypes. This multiscale modeling would enable fine and coarse analyses to observe whole system response at the different scales, and predict ideotypes for different objectives and environments.

To achieve this goal, we built a computational framework that allows asynchronous communication between models without altering their original code or timesteps. This framework allows the exchange of inputs and outputs between models, mimicking the flow of biological information across scales. We used the framework to successfully build a multiscale model of the soybean photosynthetic response to elevated CO2 concentration [CO2]. The vetted model was used to perform in silico genetic perturbations and simulate the anticipated outcomes on photosynthetic efficiency.

After dozens of simulations, we identified three top transcription factor gene candidates that will be tested experimentally for their regulatory role in controlling photosynthetic efficiency. The computational framework was also used to link whole-plant models of roots and shoots to form the core structure of virtual soybean and corn plants. These linked models were subjected to advanced visualization approaches so that in silico observations can be made on an entire stand of plants.

The use of three-dimensional architecture was shown to improve model simulations of light capture throughout a soybean canopy, providing new insights about plant competition for light resources.

The 2019 Crops in silico Symposium, Workshop & Hackathon was May 1-3 on the University of Illinois campus. After an organizing meeting for Cis 2.0 thanks to a new $5M renewing grant from the Foundation for Food and Agriculture Research, the event kicked off with a public keynote May 1, followed by a two-day symposium and workshop with the Hackathon on May 3.

Read more on the Crops in silico website >>>

 

Previous Events

The 2018 Crops in silico Symposium, Workshop & Hackathon was July 31-Aug. 3 at the Carl R. Woese Institute for Genomic Biology on the University of Illinois campus. Illinois Assistant Professor of Natural Resources and Environmental Sciences Kaiyu Guan was the keynote speaker for the three-day Symposium and Workshop; the final day was a Hackathon. The Symposium and Workshop had 55 registrants from 27 institutions and companies and nine countries. A member of the Foundation for Food and Agriculture Research (FFAR) attended the meeting for the third year in a row. Nineteen new research collaborations were formed during the workshop. Read more about the event on the Cis website >>>

The 2017 Crops in silico Symposium & Workshop was June 26-28 at the University of Oxford, UK. This second annual gathering is for experts in experimentation, agronomy, physiology, plant development, phenotyping, as well as experts in computational modeling, software development, and data visualization. The event aims to harness the great strides in understanding of plant function from genes to whole plants, to accelerate forward approaches to crop breeding and bioengineering. Read more on the Cis website >>>

The 2016 Plants in silico Symposium & Workshop — which brought together local, national, and international experts to exchange information and collaborate on a course for achieving plants in silico — was May 18-20 at the National Center for Supercomputer Applications (NCSA) on the Urbana-Champaign campus of the University of Illinois. It featured presentations by leading experts in modeling plant processes, top scientists in achieving in silico representation of other organisms, and computational scientists. Read more about the 2016 event >>>

In early 2019, the Cis project received a $5 million grant from the Foundation for Food and Agriculture Research (FFAR) to continue building a computational platform that integrates multiple models to study a whole plant virtually.

Amy Marshall-Colón, U of I Assistant Professor of Plant Biology, is the Principal Investigator for the new four-year grant. Co-Investigators include Illinois’ Matthew Turk, Assistant Professor of Astronomy and Research Scientist at the National Center for Supercomputing Applications (NCSA); Stephen P. Long, Professor of Plant Biology and Crop Sciences; Kaiyu Guan, Assistant Professor of Natural Resources and Environmental Sciences; and Meagan Lang, NCSA Research Scientist. Collaborators from other institutes include Jonathan Lynch, Professor of Plant Science at Pennsylvania State University; Bedrich Benes, Professor of Computer Graphics Technology at Purdue University; Lee Sweetlove, Professor of Plant Sciences at Oxford University; and James Schnable, Assistant Professor of Agronomy and Horticulture at the University of Nebraska.

Read the full news story >>>

In its third year, the Crops in silico team improved the accuracy of its individual and integrated plant biological models. Phenological detail was added to the data visualization. Specific highlights include:

  • The protein expression data from Stuti Shrivastava’s gene-level model was used as an input in the molecular nitrate transporter model developed by Balaji Selvam to determine nitrate uptake rate under fluctuating soil nitrate concentration.
  • Kavya Kannan improved her gene-level model of the metabolite pools of photosynthesis in response to elevated anticipated elevations of atmospheric carbon dioxide due to climate change. The model now accounts for mRNA translation and degradation rates. This model was merged with Yu Wang’s metabolic model, which contains both the dark and the light reactions of photosynthesis, to simulate leaf photosynthesis rate and metabolite concentrations under ambient and elevated CO2. Additionally, a gene regulatory network was constructed using gene expression data in response to CO2 in soybean leaves. This network can identify transcription factors regulating genes involved in photosynthesis.
  • Examples of L-systems to model plant architecture: (A) a fractal tree generated from Horton-Strahler branching patterns; (B) peach trees modeled under different water stress using L-PEACH; and (C) a photorealistic sunflower model. L-systems can be used to address a breadth of biological questions related to the evolution of plant morphology over development and in response to environmental perturbations.

    The rendering of plant- and canopy-level data derived from the system model and measured field data of soybeans over the course of a growing season was updated by AJ Christensen to include seed pod development and leaf senescence. Christensen used the geometric data both for this imagery and as input to the photosynthesis model and then rendered with colors indicating light absorption across the canopy. The data-driven plant geometry model is now implemented in an open-source tool called “L-Py” which opens up these visualization techniques to researchers without commercial graphics tools.

  • The second annual Crops in silico Symposium and Workshop was June 26-28, 2017, in Oxford. The meeting covered recent developments in plant modelling and model integration, and users and usability of models. The third annual Cis Symposium and Workshop will be July 31-Aug. 3, 2018, at the National Center for Supercomputing Applications. The objective of the event is to identify how the Cis community can communicate and coordinate their research, training and educational activities. New this year is a programming workshop and a Hackathon. At the workshop, trained Software Carpentry instructors will teach researchers basic research computing skills, which will enhance their Hackathon experience. The Hackathon will be run by Matt Turk and Meagan Lang and consist of short tutorials alternating with practical exercises that will guide participants through model integration using the prototype framework that enables communication, orchestration, and transformation, designed by Cis. The goals of the Hackathon are to improve the framework through collaboration and build social capital.

A computer simulation of how soybean plants absorb canopy light over a day.

 

In the past year, the team continued to develop and improve the accuracy of its models. Among other technical news …

  • The molecular model developed by Balaji Panneerselvam, a Chemical & Biomolecular Engineering Postdoc, was used to simulate the phosphorylated and unphosphorylated states of a key nitrate transporter and characterize the nitrate transport cycle. This model will soon be combined with the work by Stuti Shrivastava, a Plant Biology Ph.D. Candidate, whose gene-level model was optimized to determine the effects of changing nitrate uptake rate and cellular concentrations on protein levels for the nitrate transporters and their regulators. Shrivastava has also identified punitive transcription factors that can potentially regulate the expression of nitrate transporter genes.
  • Plant Biology Ph.D. Candidate Kavya Kannan improved her gene-level model of the metabolite pools of photosynthesis in response to elevated CO2 by accounting for mRNA translation and degradation rates. This model was merged with IGB Postdoc Yu Wang’s metabolic model that simulates the processes of photosynthesis under elevated CO2 and temperature and IGB Postdoc Venkat Srinivasan’s system-level model that simulates carbon partitioning between the leaf and roots, and sugar-starch portioning under elevated CO2. A manuscript reporting the result of these merged models is in preparation for submission.
  • The rendering of plant- and canopy-level data derived from the system model and measured field data of soybean over the course of a growing season was updated by AJ Christensen, a visualization programmer for NCSA’s Advanced Visualization Lab, to include seed pod development and leaf senescence.
  • Yiwen Xu, a graduate student in Computer Science, developed algorithms to reconstruct 3D plant models from video images. Her prototype system can precisely reconstruct 3D plant models from videos with little human interaction.

Matthew Turk and Amy Marshal Colón (photo by AJ Christensen, Advanced Visualization Laboratory, NCSA)

The Foundation for Food and Agriculture Research (FFAR) has awarded Principal Investigator Amy Marshall-Colón, Assistant Professor of Plant Biology at the University of Illinois at Urbana-Champaign, $274,000 to continue her research in support of Crops in silico (Cis), a project to develop a suite of virtual plant models that may help resolve a growing gap between food supply and demand in the face of global climate change.

Marshall-Colón (pictured) will collaborate with Stephen P. Long, the Gutgsell Endowed Professor of Crop Sciences and Plant Biology; Matthew Turk, Assistant Professor of Information Sciences, Assistant Research Professor of Astronomy and National Center for Supercomputing Applications (NCSA) Research Scientist (also pictured); Christine Kirkpatrick, Executive Director of the National Data Service; and Jonathan Lynch, University Distinguished Professor of Plant Science at Penn State University.

Team members will integrate above- and below-ground models of plants to create never-before-seen “whole views” of them. Then, they will subject these newly built virtual plants to computer-simulated extreme growing conditions — from flood to severe drought to increased ambient carbon dioxide — and compare the model’s predicted plant reaction to observed responses from field studies. This will help “dial in” the model’s accuracy.

The FFAR grant is the first given to a University of Illinois researcher.

Read the complete news release >>>

 

 

 

Designing crops that yield more food with fewer resource inputs is critical for future food security and sustainability. This is a tricky task for most crop breeding and engineering programs because of complex interactions between a plant’s genes, its growing environment, and farm management. New tools are needed — and soon — to help the global agricultural system meet the needs of a population climbing toward 9 billion in a climate that is rapidly changing.

 

 

In a Frontiers in Plant Science perspective article this month, a group of 30 researchers from around the world outlined a vision for a modeling framework that reconstructs a virtual plant — from the gene upward — that can accurately predict crop responses to environmental disturbances.

Crops in silico (Cis), as the envisioned tool is called, seeks a never-before-achieved level of detail, collaboration, and robustness in modeling. The goal is to create an open-source software to link individual plant models across biological scales — from a single cell to an entire ecosystem of plants.

Led by a team of plant biologists and computer scientists at the University of Illinois, more than 50 experts in modeling plant processes, computation, and informational visualization were assembled on the Urbana-Champaign campus in spring 2016 to discuss the challenges of achieving this revolutionary tool for crop research.

The Frontiers article is a synthesis of thoughts from the keynote speeches and plenary sessions at that event.

The writers acknowledge the many challenges to forming a comprehensive, integrative, internationally used modeling tool. For one, the models they hope to combine are written in a variety of coding languages and have different methods of documentation, making communication challenging. Also, many datasets that would benefit the whole modeling community are privately held.

To overcome these challenges, they highlight a four-part roadmap to success:

  1. Form a diverse and trusting research community of biologists, modelers, and visualization experts. Researchers can then more easily share data sets and work together to develop standards for models that can easily talk to one another;
  2. Build a framework capable of hosting and combining individual models. The Cis platform is intended to be the user interface from which researchers will view and launch models, libraries, data repositories, and so forth;
  3. Define a shared vocabulary, coding style, and translation guides to help unite diverse crop models into one program; and
  4. Nurture both the tool and cultivate its community of supporters and users. Annual meetings of the community, formation of a supporting nonprofit organization, and securing both federal and private funding will be essential to long-term success.

“The Cis initiative has the potential to be a powerful discovery tool in which dozens of simulations across multiple scenarios can be accomplished in a few hours,” the authors wrote.

Beyond a technological breakthrough, the Cis team also aims to achieve a research community shift.

“We believe Crops in silico will unite largely isolated efforts into a connected and collaborative community that can take full advantage of advances in computation science and mechanistic understanding of plant processes and their responses to the environment,” said lead author and Plant Biology Assistant Professor Amy Marshall-Colón, a Principal Investigator in the Crops in silico project at Illinois.

The Cis project was seed-funded in 2015 by a $350,000 grant from the Institute for Sustainability, Energy, and Environment (iSEE) at the University of Illinois at Urbana-Champaign. The Crops in silico symposium was supported by iSEE, the National Center for Supercomputing Applications (NCSA), the Carl R. Woese Institute for Genomic Biology (IGB), the College of Agricultural, Consumer and Environmental Sciences, the School of Molecular and Cellular Biology, and the Departments of Crop Sciences and Plant Biology and the Olga G. Nalbandov Lecture Funds. Read more on the project page on iSEE’s website, or visit the Cis homepage at cropsinsilico.org.

To see the full perspective article on Cis, visit the Frontiers in Plant Science website.

 

 

Cis co-PI Amy Marshall-Colón and team member AJ Christensen, a National Center for Supercomputing Applications (NCSA) Visualization Programmer, attended the Global Open Data For Agriculture and Nutrition (GODAN) Summit in September 2016 in New York. Marshall-Colón and Christensen created a writeup on the summit and a mini-documentary on the power of visualization and digital labs in helping to solve the impending world food crisis.

 

 

Read more on the NCSA website >>>

Rendered plant- and canopy-level data from the Psi system-level model.

Rendered plant- and canopy-level data from the Cis system-level model.

 

In the past year, some technical developments …

  • Cis has developed multiple models:
    1. a molecular model of At1, a key nitrate transporter;
    2. a gene-level model of the effects of changing nitrate uptake rate and cellular concentrations on transcript and metabolite levels;
    3. a gene-level model of the effects of changing atmospheric carbon dioxide concentrations on transcript and metabolite levels;
    4. a metabolic model that simulates the processes of photosynthesis under elevated atmospheric CO2 concentrations;
    5. a system-level model that simulates carbon partitioning between the leaf and roots, and sugar-starch portioning under elevated atmospheric CO2 concentrations;
    6. a solar light absorption model using bi-directional path tracing.

      Photo by Haley Ahlers Data for the system-level model was derived from soybean trials on the University of Illinois South Farms.

      Photo by Haley Ahlers
      Data for the system-level model was derived from soybean trials on the University of Illinois South Farms.

  • Cis has rendered plant- and canopy- level data derived from the system model and measured field data of soybean over the course of a growing season. This data includes phenotypic differences between soybean growing under ambient and elevated atmospheric CO2.
  • Cis constructed a run-time communications system for integration. The system provides global control, logging, and debugging support and permits new models to be integrated without requiring change to existing models. The metabolic and bio-system models are on line, with others being added as they become ready for testing.


The Cis 2.0 Team

Front row, from left: Kaiyu Guan, Bedrich Benes, Amy Marshall-Colón, Stephen P. Long, and Traci Quigg Thomas. Back row: James Schnable, Meagan Lang, Lee Sweetlove, and Matthew Turk.

Principal Investigators and co-PIs (University of Illinois unless otherwise noted)

 

Operating Team


Project Manager

 

Former co-PIs and Team Members

 

Former Project Advisers


Publications & Presentations

(iSEE project members’ names in bold)

  • Publication: “Combining Gene Network, Metabolic, and Leaf-Level Models Show Means to Future-Proof Soybean Photosynthesis under Rising CO2.” Kannan, K.; Wang, Y.; Lang, M.; Challa, G.S.; Long, S.P.; Marshall-Colón, A. in silico Plants 1,1. June 2019.
  • Presentation: “Current Status of the Crops in silico Framework.” Lang, M. Fourth Annual Cis Symposium & Workshop, Urbana, Ill., May 2019.
  • Presentation: Hackathon Tutorial on How to Use the yggdrasil Package. Cis team. Fourth Annual Cis Symposium & Workshop, Urbana, Ill., May 2019.
  • Publication: “yggdrasil: A Python Package for Integrating Computational Models across Languages and Scales.” Lang, M. in silico Plants 1,1. April 2019.
  • Presentation: “Multiscale Modeling and Visualization Platform for Crops.” Wang, Y. Third Annual Cis Symposium & Workshop, Urbana, Ill., August 2018.
  • Presentation: “Integrative Modeling and Visualization for the Development of in Silico Crops.” Marshall-Colón, A. Plant Biology 2018, Montreal, Quebec, July 18, 2018.
  • Presentation: “Cis_Interface: A Python Package for Connecting Scientific Models across Scales and Languages.” Lang, M. SciPy 2018, Austin, TX, July 11, 2018.
  • Presentation: “Modeling Dependence in Signaling Cascades using Dynamic Transcriptome Data.” Marshall-Colón, A. Interdisciplinary Plant Group: Advances in Plant Metabolism, Columbia, Mo., June 1, 2018.
  • Publication: “Use of Computational Modeling Combined with Advanced Visualization to Develop Strategies for the Design of Crop Ideotypes to Address Food Security.” Christensen, A.J.Srinivasan, V.Hart, J.C.Marshall-Colón, A. Nutrition Reviews, May 2018.
  • Presentation: “Advanced Visualization for Crops in silico.” Christensen, A.J. School of Information Sciences, University of Illinois, Urbana, Ill., March 12, 2018.
  • Presentation: “Revealing the Conformational Flexibility of Dual Affinity Nitrate Transporter (NRT1.1) Gene in Arabidopsis Thaliana.” Panneerselvam, B.Shukla, D. Monsanto Research Symposium (best poster award), Urbana, Ill., June 30, 2017.
  • Presentation: “Developing an Integration Framework.” Lang, M. Second Annual Crops in silico Symposium and Workshop, Oxford, UK, June 27, 2017
  • Publication: “Crops in silico: Generating Virtual Crops Using an Integrative and Multi-scale Modeling Platform.” Marshall-Colón, A.; Long, S.P; Allen, D.K.; Allen, G.; Beard, D.A.; Benes, B.; von Caemmerer, S.; Christensen, A.J.; Cox, D.J.; Hart, J.C.; Hirst, P.M.; Kannan, K.; Katz, D.S.; Lynch, J.P; Millar, A.J.; Panneerselvam, B.;  Price, N.D.; Prusinkiewicz, P.; Raila, D.; Shekar, R.G.; Shrivastava, S.; Shukla, D.; Srinivasan, V.; Stitt, M.; Turk, M.J.; Voit, E.O.; Wang, Y.; Yin, X.; Zhu, X. Frontiers in Plant Science, May 2017.
  • Presentation: “Understanding the Conformational Switches in Membrane Transporter Proteins Using Blue Waters.” Panneerselvam, B.; Mittal, S; Shukla, D. Blue Waters Symposium, Sunriver, Ore., May 16-19, 2017.
  • Presentation: “Automistic Determination of the Conformational Switches in Membrane Transporter Proteins.” Panneerselvam, B.; Shukla, D. 2017 Blue Waters Symposium, Sunriver, Ore., May 2017.
  • Presentation: “Understanding the Conformational Diversity of the Proton-coupled Oligopeptide Transporter (POT) Family.” Panneerselvam, B. 61st Annual Meeting of the Biophysical Society, New Orleans, La., February 2017.
  • Presentation: “Crops in silico: A Communitywide Multi-scale Computational Modeling Framework of Plant Canopies.” Srinivasan, V.; Christensen, A.J.; Borkiewicz, K.; Xu, Y.Ellis, A.; Panneerselvam, B.; Kannan, K.; Shrivastava, S.; Cox, D.J.; Hart, J.; Marshall-Colón, A.; Long, S.P. American Geophysical Union Fall Meeting, San Francisco, Calif., December 2016.
  • Presentation: “A Vision of Crops in silico – from Gene Function to a Growing Plant.” Marshall-Colón, A. Crop Engineering Consortium, London, England, October 2016.
  • Presentation: “Research Discussion — Crops in silico.” Marshall-Colón, A. Grand Challenges Annual Meeting, London, England, October 2016.
  • Presentation and video: How Supercomputers and Visualization Can Help the Global Food Crisis,” Global Open Data for Agriculture and Nutrition (GODAN) Summit, New York, N.Y., September 2016.
  • Presentation: “In-silico Model Integration of Nitrogen Transporter, NRT1.1 with Nitrogen Responsive Transcription Factor Network,” Shrivastava, S. Poster session at the Plant Biology 2016 Meeting in Austin, Texas, July 2016.
  • Presentation: “In-silico Predictions of Conformational States of Plant and Bacterial Transporters,” Panneerselvam, B., Mittal, S., Shukla, D. Poster session at the Plant Biology 2016 Meeting in Austin, Texas, July 2016.
  • Publication: “Crops in silico: Why, Why Now and What? An Integrative Platform for Plant Systems Biology Research,” Zhu, X.; Lynch, J.P.; LeBauer, D.S.; Millar, A.J.; Stitt, M.; Long, S.P. Plant, Cell & Environment, January 2016.

Read More about Cis …

Follow Online

Website: cropsinsilico.org

X