An introduction to whole-cell modeling

Despite extensive research into individual molecules and pathways, we still do not have a complete understanding of cell biology. To comprehensively understand cells, we must build whole-cell (WC) computational models that predict phenotype from genotype by representing all of the biochemical activity inside cells. In addition to helping researchers gain novel insights into biology, WC models could also help bioengineers design microorganisms and help physicians personalize medicine. Recently, we and others demonstrated that WC models are feasible by leveraging recent advances in computational and experimental technology to build the first model that represents every characterized gene in a cell. Although substantial work remains to develop comprehensive WC models, we believe that WC models can be achieved by systemizing cell modeling and coordinating the cell modeling community.

The goals of this primer are to teach researchers about the motivation, goals, principles, and methods of WC modeling and to prepare researchers to contribute to WC modeling. Toward this goal, the primer includes introductory readings and hands-on tutorials on WC modeling, the fundamental principles and methods of cell modeling, computer programming and software engineering, Linux computer systems, and scientific communication. Section 1 introduces the motivation, goals, and fundamental challenges of WC modeling; describes why WC modeling is becoming feasible; summarizes the principles and methods of WC modeling; reviews the latest WC models and their limitations; outlines the major bottlenecks to WC modeling; proposes a plan for achieving WC models as a community; and describes ongoing work to advance WC modeling. Section 2 introduces several foundational concepts and skills for WC modeling including computer programming, Linux computer systems, and scientific communication. Section 3 outlines the theoretical foundations for WC modeling. Section 4 describes the principles and methods of WC modeling. Section 5 outlines several funding opportunities for WC modeling.

Please note, this primer is under active development. Over time, we aim to develop a complete primer with detailed examples and tutorials. We welcome suggestions and feedback.

• 1. Introduction
  • 1.1. Motivation for WC modeling
    • 1.1.1. Biological science: understand how genotype influences phenotype
    • 1.1.2. Medicine: personalize medicine for individual genomes
    • 1.1.3. Synthetic biology: rationally design microbial genomes
  • 1.2. The biology that WC models should aim to represent and predict
    • 1.2.1. Phenotypes that WC models should aim to predict
    • 1.2.2. Physics and chemistry that WC models should aim to represent
  • 1.3. Fundamental challenges to WC modeling
    • 1.3.1. Integrating molecular behavior to the cell level over several spatiotemporal scales
    • 1.3.2. Assembling a unified molecular understanding of cells from imperfect data
    • 1.3.3. Selecting, calibrating and validating high-dimensional models
  • 1.4. Feasibility of WC models
    • 1.4.1. Experimental methods, data, and repositories
    • 1.4.2. Modeling and simulation tools
    • 1.4.3. Models of individual pathways and model repositories
    • 1.4.4. Models of multiple pathways
  • 1.5. Emerging principles and methods for WC modeling
    • 1.5.1. Principles of WC modeling
    • 1.5.2. Methods for WC modeling
  • 1.6. Latest WC models and their limitations
    • 1.6.1. Coarse-grained models
    • 1.6.2. Genomically-centric bottom-up fine-grained models
    • 1.6.3. Physiologically-centric top-down fine-grained models
    • 1.6.4. Spatially-centric bottom-up fine-grained models
    • 1.6.5. Hybrid models
  • 1.7. Bottlenecks to more comprehensive and predictive WC models
    • 1.7.1. Inadequate experimental methods and data repositories
    • 1.7.2. Incomplete, inconsistent, scattered, and poorly annotated pathway models
    • 1.7.3. Inadequate software tools for WC modeling
    • 1.7.4. Inadequate model formats
    • 1.7.5. Lack of coordination among the cell modeling community
  • 1.8. Technologies needed to advance WC modeling
    • 1.8.1. Experimental methods for characterizing cells
    • 1.8.2. Tools for aggregating, standardizing, and integrating heterogeneous data
    • 1.8.3. Tools for scalably designing models from large datasets
    • 1.8.4. Rule-based format for representing models
    • 1.8.5. Scalable network-free, multi-algorithmic simulator
    • 1.8.6. Scalable tools for calibrating models
    • 1.8.7. Scalable tools for verifying models
    • 1.8.8. Additional tools that would help accelerate WC modeling
  • 1.9. A plan for achieving comprehensive WC models as a community
    • 1.9.1. Phase I: Piloting the core technologies and concepts of WC modeling
    • 1.9.2. Phase II: Piloting collaborative WC modeling
    • 1.9.3. Phase III: Community modeling and model validation
  • 1.10. Ongoing efforts to advance WC modeling
    • 1.10.1. Genomically-centric models
    • 1.10.2. Physiologically-centric, spatially-centric, and hybrid models
    • 1.10.3. Technology development
  • 1.11. Resources for learning about WC modeling
    • 1.11.1. Summer schools
    • 1.11.2. Online forum
  • 1.12. Outlook
• 2. Foundational concepts and skills for computational biology
  • 2.1. Typing
  • 2.2. Software engineering
    • 2.2.1. An introduction to Python
    • 2.2.2. Numerical computing with NumPy
    • 2.2.3. Plotting data with matplotlib
    • 2.2.4. Developing database, command line, and web-based programs with Python
    • 2.2.5. Writing code for Python 2 and 3
    • 2.2.6. Organizing Python code into functions, classes, and modules
    • 2.2.7. Structuring Python projects
    • 2.2.8. Revisioning code with Git, GitHub, and Meld
    • 2.2.9. Testing Python code with unittest, pytest, and Coverage
    • 2.2.10. Debugging Python code using the PyCharm debugger
    • 2.2.11. Documenting Python code with Sphinx
    • 2.2.12. Continuously testing Python code with CircleCI, Coveralls, Code Climate, and the Karr Lab’s dashboards
    • 2.2.13. Distributing Python software with GitHub, PyPI, Docker Hub, and Read The Docs
    • 2.2.14. Recommended Python development tools
    • 2.2.15. Comparison between Python and other languages
  • 2.3. Linux
    • 2.3.1. How to build a Linux Mint virtual machine with Virtual Box
    • 2.3.2. How to build a Ubuntu Linux image with Docker
    • 2.3.3. An introduction to Linux Mint
  • 2.4. Version and sharing data with Quilt
    • 2.4.1. Overview
    • 2.4.2. Using Quilt
  • 2.5. Scientific communication: papers, presentations, graphics
    • 2.5.1. Writing manuscripts
    • 2.5.2. Reviewing manuscripts
    • 2.5.3. Making posters
    • 2.5.4. Making presentations
    • 2.5.5. Visualizing data
    • 2.5.6. Formatting textual documents with LaTeX
    • 2.5.7. Drawing vector graphics with Adobe Illustrator and Inkscape
    • 2.5.8. Editing raster graphics with Gimp
• 3. Fundamentals of cell modeling
  • 3.1. Data aggregation
    • 3.1.1. Common data types and data sources
    • 3.1.2. Finding data sources
    • 3.1.3. Finding relevant data for models
    • 3.1.4. Data aggregation tools
    • 3.1.5. Determining the consensus of multiple observations
    • 3.1.6. Exercise
  • 3.2. Input data organization
    • 3.2.1. Schema
    • 3.2.2. Software tools
    • 3.2.3. Exercises
  • 3.3. Model design
    • 3.3.1. Software tools
    • 3.3.2. Exercises
  • 3.4. Model calibration
    • 3.4.1. Key concepts
    • 3.4.2. Calibration data
    • 3.4.3. Approximate, multi-stage parameter estimation
    • 3.4.4. Exercise
  • 3.5. Model representation
    • 3.5.1. Custom numerical simulation code
    • 3.5.2. Standard numerical simulation packages
    • 3.5.3. Enumerated modeling languages
    • 3.5.4. Ruled-based modeling languages
    • 3.5.5. Rule-based modeling API
    • 3.5.6. High-level rule-based modeling language
  • 3.6. Model annotation
    • 3.6.1. Component-level semantic annotations of species, reactions, and parameters
    • 3.6.2. Model-level semantic annotations
    • 3.6.3. Model provenance annotations
    • 3.6.4. Simulation algorithm annotations
    • 3.6.5. Exercises
  • 3.7. Model composition
    • 3.7.1. Model composition procedure
    • 3.7.2. Software tools
    • 3.7.3. Exercises
  • 3.8. Mathematical representations and simulation algorithms
    • 3.8.1. Boolean/logical models
    • 3.8.2. Ordinary differential equations (ODEs)
    • 3.8.3. Stochastic simulation
    • 3.8.4. Network-free simulation
    • 3.8.5. Flux balance analysis (FBA)
    • 3.8.6. Hybrid/multi-algorithmic simulation
    • 3.8.7. Reproducing stochastic simulations
    • 3.8.8. Simulation descriptions
    • 3.8.9. Software tools
    • 3.8.10. Exercises
  • 3.9. Model testing
    • 3.9.1. Testing composite models
    • 3.9.2. Exercise
  • 3.10. Logging simulation results
  • 3.11. Organizing simulation results
    • 3.11.1. Exercise
  • 3.12. Quickly analyzing large simulation results
  • 3.13. Rule-based Modeling with BioNetGen, BNGL, and RuleBender
    • 3.13.1. Description of Rule-based Modeling, BioNetGen, BNGL, and RuleBender
    • 3.13.2. BNGL and RuleBender Basic Functionality
    • 3.13.3. Exercises
    • 3.13.4. Molecular sites, their states, and bonds
• 4. Principles and methods of WC modeling
  • 4.1. Units of WC models
  • 4.2. Using the wc_lang package to define whole-cell models
    • 4.2.1. Semantics of a wc_lang biochemical Model
    • 4.2.2. wc_lang Classes Used to Define biochemical Models
    • 4.2.3. Using wc_lang
  • 4.3. Using wc_env_manager build, version, and sharing computing environments for WC modeling
    • 4.3.1. How wc_env_manager works
    • 4.3.2. Installing wc_env_manager
    • 4.3.3. Using wc_env_manager to build and share images for WC modeling
    • 4.3.4. Using wc_env_manager to create and run Docker containers for WC modeling
    • 4.3.5. Using WC modeling computing environments with an external IDE such as PyCharm
    • 4.3.6. Caveats and troubleshooting
• 5. Appendix: Funding WC modeling research
  • 5.1. Graduate fellowships
  • 5.2. Postdoctoral fellowships
  • 5.3. Postdoc/faculty transition awards
  • 5.4. Grants
    • 5.4.1. Funding streams for your lab
    • 5.4.2. Taxonomy of funding opportunities
    • 5.4.3. Funding programs for early career investigators
    • 5.4.4. Finding funding opportunities
    • 5.4.5. Eligibility
    • 5.4.6. Deadlines
    • 5.4.7. Proposal process
    • 5.4.8. Writing proposals
    • 5.4.9. Typical costs for budgets
    • 5.4.10. Submitting proposals
    • 5.4.11. Peer review
    • 5.4.12. Statistics (NIGMS)
    • 5.4.13. Grant award process
    • 5.4.14. Annual grant renewals
    • 5.4.15. Advice for winning grants
    • 5.4.16. Advice for resubmissions
• 6. Appendix: Installing the code in this primer
  • 6.1. Requirements
  • 6.2. How to install these tutorials
  • 6.3. Detailed instructions to install the tutorials and all of the requirements
• 7. Appendix: Documentation for the code in this primer
  • 7.1. Subpackages
    • 7.1.1. intro_to_wc_modeling.cell_modeling package
    • 7.1.2. intro_to_wc_modeling.concepts_skills package
    • 7.1.3. intro_to_wc_modeling.wc_modeling package
  • 7.2. Submodules
  • 7.3. intro_to_wc_modeling._version module
  • 7.4. Module contents
• 8. Appendix: Acronyms
• 9. Appendix: Glossary
• 10. Appendix: References
• 11. Appendix: About this primer
  • 11.1. License
  • 11.2. Authors
  • 11.3. Acknowledgements
  • 11.4. Questions and comments