Computational Modeling of Signaling Networks

Preface
Contents
Contributors
Part I: Advances in Computational Modelling of Signalling Networks
Chapter 1: Design Principles Underlying Robust Adaptation of Complex Biochemical Networks
1 Introduction
2 The Mathematics of RPA
2.1 Early Models of RPA in Biology
2.2 The Mathematical Basis for RPA in General Interaction Networks
3 Methods for RPA Analysis
3.1 A Diagrammatic Method for Analyzing the RPA Capacity of a Network
3.2 Fundamental Design Principles of RPA Networks
4 Applications of the Diagrammatic Method to Example Networks
4.1 Example Network 1: An Opposer Module in Parallel with a Balancer Module
4.2 Example Network 2: An Opposer Module in Series with a Balancer Module
4.3 Example Network 3: A Single Opposer Module with Two-Node Opposing Set and a Single Input-Output Node
4.4 Example Network 4: A Single RPA-Capable Network with Two Possible RPA Solutions
4.5 Example Network 5: A 12-Node RPA-Capable Network with Two Opposer Modules in Parallel
5 Concluding Remarks: Future Directions for RPA Theory
References
Chapter 2: Multi-Dimensional Analysis of Biochemical Network Dynamics Using pyDYVIPAC
1 Introduction
2 Overview of pyDYVIPAC
3 Requirements for Running pyDYVIPAC
3.1 Installation Requirements
3.2 Description of Key Input Files for pyDYVIPAC
4 Running pyDYVIPAC
4.1 Setting Up and Running pyDYVIPAC Jupyter Notebook
4.2 Displaying and Analyzing pyDYVIPAC Outputs
5 Demonstration of pyDYVIPAC Using Specific Examples
5.1 Example 1: The Negative Feedback Goodwin System
5.1.1 Setting Up
5.1.2 Parameter Sampling and Dynamics Processing
5.1.3 Visualization of Results
5.1.4 Explore the Hyperspace in Higher Dimensions
5.2 Example 2: The Mixed Feedback Phosphorylation Cascade
6 Summary
7 Notes
References
Chapter 3: A Practical Guide for the Efficient Formulation and Calibration of Large, Energy- and Rule-Based Models of Cellular...
1 Introduction
2 Mathematical Problem Formulation
3 Formulating a Thermodynamic Model of RAF Inhibition in PySB
3.1 Protein Species
3.2 Protein Interactions
3.3 RAF Inhibitor
3.4 Paradoxical Activation
4 Importing Data in PEtab Format
4.1 Observables
4.2 Measurements and Conditions
4.3 Parameters
4.4 SBML Export
5 Calibrating the Model in pyPESTO
6 Discussion
References
Chapter 4: Systems Biology: Identifiability Analysis and Parameter Identification via Systems-Biology-Informed Neural Networks
1 Introduction
2 Ultradian Endocrine Model for Glucose-Insulin Interaction
3 Structural Identifiability Analysis
3.1 Preprocessing
3.2 Structural Identifiability Results
4 Parameter Estimation via SBINN
4.1 Deep Neural Networks
4.2 Systems-Biology-Informed Neural Networks (SBINN)
4.3 Results of SBINN
5 Practical Identifiability Analysis
6 Discussion of Time-Dependent Parameters
7 Summary
Appendix
A. Python
B. Julia
References
Chapter 5: A Practical Guide to Reproducible Modeling for Biochemical Networks
1 Introduction
2 Data Collection
2.1 Databases for Modeling Biochemical Networks
2.2 Metadata Management
2.3 Automated Data Collection from Online Databases
3 Model Construction
3.1 Structured Model Description Formats
3.2 Understanding the System Through Visualization
3.3 Naming Conventions and Annotation
3.4 Documentation and Versioning
4 Simulation
4.1 Writing a Simulation Experiment
4.2 Explicit Representation of Simulation Inputs, Algorithms, and Initial Conditions
4.3 Executing a Simulation Experiment
4.4 Recording Simulation Results
5 Parameter Estimation
5.1 Parameter Estimation Algorithm and Reporting Estimated Parameter Values
5.2 Uncertainty Quantification and Reporting Parameter Distributions with Monte Carlo Sampling
6 Model Testing
6.1 Automated Model Verification and Validation Using a Test Suite
6.2 Replication in an Independent Computing Environment to Check Portability
7 Packaging and Publishing
7.1 Standalone Packaging
7.2 Model Repositories
8 Discussion of Guiding Principles
8.1 FAIR Principles
8.2 Standardized Formats
8.3 Model Versioning
8.4 Automated Testing, Version Control, Containerized Solutions, and Continuous Integration
9 Conclusions
References
Chapter 6: Integrating Multi-Omics Data to Construct Reliable Interconnected Models of Signaling, Gene Regulatory, and Metabol...
1 Introduction
2 Materials
2.1 Dataset
2.1.1 Dataset for Knowledgebase Meta-Interactome Framework Development
2.1.2 Dataset for Multi-Omics Data Analysis
2.1.3 B. Microbiome Data
3 Methods
3.1 Meta-Interactome Framework Development
3.1.1 Construction of Meta-Interaction Networks
3.2 Multi-Omics Data Analysis
3.2.1 Differential Expression Analysis
3.2.2 Microbiome Data Processing and Analysis
3.2.3 Reconstruction of Metagenome Content from Amplicon Data
3.2.4 Estimation of Community-Wise Metabolic Potential
3.2.5 Correlation Analysis Between Predicted and User-Supplied Metabolites
3.3 Multi-Omics Data Integration and Meta-Interaction Network Extraction
3.3.1 Identification of Topologically Important Nodes (TINs)
3.3.2 Metabolic Enzyme Cross-Connecting Paths and Networks
3.3.3 Contextualization and Customization of the Cross-Connecting Networks
3.3.4 Identification of Significant Interconnecting Pairs and Paths
3.3.5 In Silico Perturbation Analysis
4 Conclusions
References
Chapter 7: Efficient Quantification of Extrinsic Fluctuations via Stochastic Simulations
1 Introduction
2 Nanog Transcriptional Regulatory Network as Model System
3 Method
4 Applications
5 Conclusion
References
Chapter 8: Meta-Dynamic Network Modelling for Biochemical Networks
1 Introduction
2 Overview of Meta-Dynamic Network (MDN) Modelling
3 Case Study: The Hippo-ERK Crosstalk Network
4 Application of MDN Modelling to the Hippo-ERK Crosstalk Network
4.1 MDN Setup and Implementation
4.2 MDN Modelling-Based Analysis of Protein Expression Heterogeneity
4.3 Identification of Parameter Patterns Controlling Particular pERK Dynamics
5 Discussion
Appendix A: Ordinary Differential Equations of the MST2-ERK Crosstalk Model
Appendix B: Reactions and Reaction Rates of the MST2-ERK Network Model
Appendix C: Parameter Values of the MST2-ERK Network Model
References
Chapter 9: Rapid Particle-Based Simulations of Cellular Signalling with the FLAME-Accelerated Signalling Tool (FaST) and GPUs
1 Introduction
2 Materials
2.1 FaST Installation
2.2 Building, Compilation, and Running FaST Models for GPUs
3 Methods
3.1 Creating a Simulation with FaST
3.1.1 Writing the Agents File
3.1.2 Writing the Reactions File
3.1.3 Running FaST
3.1.4 Evaluating the Code
3.2 Compilation with FLAME GPU
3.2.1 Creating a New Project
3.2.2 Setting Up Microsoft Visual Studios (Windows Only)
3.2.3 Compilation with Microsoft Visual Studios (Windows Only)
3.2.4 Compilation with Make (Unix Only)
3.3 Running a Simulation
3.3.1 Creating the Initial States
3.3.2 Running a Simulation
3.3.3 Analyzing the Data
3.3.4 Adding a Visualization
3.4 Customizing a Simulation
3.4.1 Adding an Entirely New Agent
3.4.2 Adding and Modifying a Function
3.4.3 Modifying the Initial States
4 Notes
References
Part II: Advances in Integrative Analysis of Signalling Networks
Chapter 10: Modeling Cellular Signaling Variability Based on Single-Cell Data: The TGFβ-SMAD Signaling Pathway
1 Introduction
1.1 Heterogeneity in Cellular Responses to External Cues
1.2 Heterogeneity in the Activity of Cellular Signaling Pathways
1.3 Signaling Fluctuations Are Often Nongenetic and Temporally Stable
1.4 Signaling Heterogeneity Arises from Fluctuations in Signaling Protein Levels
2 Methods
2.1 Mathematical Modeling of Cellular Heterogeneity
2.1.1 Applications and Limitations of Population-Average Models
2.1.2 Deterministic Models of Signaling Heterogeneity: Implementation and Scope
2.1.3 Stochastic Modeling of Signaling and Gene Expression Heterogeneity
2.1.4 Quantitative Modeling of Cellular Heterogeneity
2.2 Heterogeneity in TGFβ Signaling: Modeling and Impact on Cellular Behavior
2.2.1 TGFβ Signaling in Health and Disease
2.2.2 Lessons Learned from Single-Cell Experiments of TGFβ Signaling
2.2.3 Population-Average Models of TGFβ/SMAD Signaling
2.2.4 Towards Quantitative Modeling of SMAD Signaling Heterogeneity
2.2.5 Future Directions: Link Between Signaling and Gene Expression Heterogeneity
References
Chapter 11: Quantitative Imaging Analysis of NF-κB for Mathematical Modeling Applications
1 Introduction
2 Materials
2.1 Cell Culture
2.2 Microscopy and Image Acquisition Software
2.3 Image and Data Analysis
3 Methods
3.1 Image Analysis Using FIJI-ImageJ
3.2 Procedure for Sample Image Analysis
3.3 Data Analysis Using R and RStudio
4 Note
5 Conclusions
References
Chapter 12: Resolving Crosstalk Between Signaling Pathways Using Mathematical Modeling and Time-Resolved Single Cell Data
1 Introduction
2 Extracting Information About Crosstalk from Altered Signaling Response
3 Model Selection and Parameterization
3.1 Clustering of Single Cell Trajectories
3.2 Subpopulation-Specific Modeling
4 Identifying Points of Interaction Between Two Signaling Pathways
4.1 Sensitivity Analysis-Based Prediction
4.2 Parameter Inference-Based Prediction
5 Experimental Validation of Model-Based Predictions
6 Conclusion
References
Chapter 13: Live-Cell Sender-Receiver Co-cultures for Quantitative Measurement of Paracrine Signaling Dynamics, Gene Expressio...
1 Introduction
2 Materials
2.1 Cell Lines
2.2 Cell Culture Reagents
2.3 CRISPR Tagging and Validation Reagents
2.4 Genetically Encoded Reporter Constructs
2.5 Viral Production Reagents
2.6 Live Cell and Fixed Imaging Materials and Reagents
2.7 Live Cell Microscopy Equipment
2.8 Image Processing and Modeling
3 Methods
3.1 CRISPR Gene Tagging
3.2 Validation of CRISPR Gene Editing
3.3 Establishing Reporter Cell Lines
3.4 Imaging Experiment Preparation
3.5 Validation of Reporter Cells
3.6 Establishing Co-culture Conditions
3.7 Image Processing
3.8 Analysis of Co-culture Signaling and Gene Expression Responses
4 Notes
References
Chapter 14: Application of Optogenetics to Probe the Signaling Dynamics of Cell Fate Decision-Making
1 Introduction
2 Materials
3 Methods
3.1 Calibrating of Optogenetic Tools
3.2 Determining the Minimum Signaling Dynamics for Cell Fate
4 Notes
References
Part III: Application of Integrative Modelling and Analysis of Signalling Networks in Diseases
Chapter 15: Computational Random Mutagenesis to Investigate RAS Mutant Signaling
1 Introduction
2 Materials
2.1 RAS Model
2.2 Parameter Variation Plan for Computational Random Mutagenesis
2.3 Software to Implement the Model and Hardware for Simulations
3 Methods
4 Notes
References
Chapter 16: Mathematically Modeling the Effect of Endocrine and Cdk4/6 Inhibitor Therapies on Breast Cancer Cells
1 Introduction
2 Material
3 Methods
3.1 Signaling Wiring Diagram and Model Structure
3.2 Model Implementation and Parameter Estimation
3.3 Local Sensitivity Analysis
4 Notes
4.1 Results
4.2 Discussion
References
Chapter 17: SynDISCO: A Mechanistic Modeling-Based Framework for Predictive Prioritization of Synergistic Drug Combinations Ta...
1 Introduction
2 Description of SynDISCO
2.1 Module 1: ODE Model Construction
2.2 Predictive Prioritization of Drug Combinations
2.3 Experimental Validation of Top Candidates
3 Application of SynDISCO to the EGFR-PYK2 Signaling Network in TNBC
3.1 Construction of the EGFR-MET Network Model
3.2 Drug Response Simulation of the EGFR-MET Model
3.3 Experimental Validation of Top-Predicted Drug Combinations
4 Summary
5 Notes
Appendix A. Ordinary Differential Equations of the EGFR-MET Model
Appendix B. Reactions and Reaction Rates of the EGFR-MET Network Model
Appendix C. Best-Fitted Parameter Values Used for Simulations
References
Index