**Model Selection and Network Inference of Complex Systems**

**複雜系統的模型選擇及網路推論**

#### Course description

This course will make you better at building mathematical models of systems. Many different types of relationships among variables can be represented as graphs where links connect nodes, e.g. control systems, communication networks, stress in frames of buildings, and gene regulation.

In Network inference such graphs are identified based on quantitative changes in the variables. When building a model of a system, one needs to select which model to use and which features to include in it. All statements made based on the final model are only valid under the assumption that the model structure and feature selection is correct. Model selection is therefore essential in system identification.

This course will give you an understanding of the network inference and feature selection problem, when and how to use network inference feature selection. You will also get hands-on experience applying network inference and feature selection to a data set of your choosing.

#### Course Objectives

- Search for articles in some scientific reference database.
- Write a summary of an article in relation to a specific question.
- Write a technical report.
- Critically review another person’s text.
- Explain the network inference problem.
- Explain the feature selection problem.
- Explain the relation between network inference and feature selection.
- Explain what model selection is and why it is essential in system identification
- Illustrate relations among variables graphically.
- Illustrate feature selection visually.
- Evaluate different feature selection or network inference algorithms.
- Apply some feature selection or network inference algorithm to a data set.
- Explain the difference between data, information, and knowledge.
- Analyse the quality of an inferred network.
- Summarize the current state of the art in network inference.
- Summarize the current state of the art in feature selection.
- Implement a feature selection or network inference algorithm in Matlab or Python.

#### Teaching Strategies

How this course is going to be taught

- Lecture 100%

#### Course material

The material will mainly consist of recent published peer-reviewed articles on network inference, feature selection, and model selection.

##### Course outline

- Introduction to modelling
- Introduction to model selection
- Analysis of model selection
- Introduction to feature selection
- Analysis of Feature selection
- Introduction to network inference
- Introduction to machine learning
- Overview of the state of the art in feature selection
- Analysis of the state of the art in feature selection
- Overview of the state of the art in network inference Overview of the state of the art in network inference
- Analysis of the state of the art in network inference
- Application of feature selection or network inference
- Implementation of feature selection or network inference algorithms
- Implementation of feature selection or network inference algorithms
- Implementation of feature selection or network inferencealgorithms
- Analysis of the quality of selected features/models
- Evaluation of feature selection or network inference algorithms

##### References

- Hecker, M., Lambeck, S., Toepfer, S., van Someren, E. & Guthke, R. Gene regulatory network inference: data integration in dynamic models-a review. Biosystems. 96, 86–103 (2009).
- De Smet, R. & Marchal, K. Advantages and limitations of current network inference methods. Nat. Rev. Microbiol. 8, 717–29 (2010).
- De Jong, H. Modeling and simulation of genetic regulatory systems: a literature review. J Comput Biol 9, 67–103 (2002).
- Emmert-Streib, F., Glazko, G. V, Altay, G. & de Matos Simoes, R. Statistical inference and reverse engineering of gene regulatory networks from observational expression data. Front. Genet. 3, 8 (2012).
- Markowetz,F. & Spang, R. Inferring cellular networks–a review. BMC Bioinformatics 8 Suppl 6, S5 (2007).
- Bonneau, R. Learning biological networks: from modules to dynamics. Nat. Chem. Biol. 4, 658–64 (2008).
- He, F., Balling, R. & Zeng, A.-P. Reverse engineering and verification of gene networks: principles, assumptions, and limitations of present methods and future perspectives. J. Biotechnol. 144, 190–203 (2009).
- Guyon, I. & Elisseeff, A. An Introduction to Variable and Feature Selection. J. Mach. Learn. Res. 3, 1157–1182 (2003).
- Batanlar, Y. & Özuysal, M. Introduction to machine learning. Methods Mol. Biol. 1107, 105–128 (2014).
- Fan, J. & Lv, J. A Selective Overview of Variable Selection in High Dimensional Feature Space (Invited Review Article). Stat. Sin. 20, 101–148 (2010).
- George, E. I. The variable selection problem. J. Am. Stat. Assoc. 95, 1304–1308 (2000).
- Stoica, P. & Selen, Y. Model-order selection: a review of information criterion rules. IEEE Signal Process. Mag. 21, 36–47 (2004).
- Nordling, T. E. M. Robust inference of gene regulatory networks: System properties, variable selection, subnetworks, and design of experiments. (2013).
- Gross, J. L. & Yellen, J. Graph theory and its applications. (CRC Press, 1999).
- Albert, R. & Barabási, A.-L. Statistical mechanics of complex networks. Rev. Mod. Phys. 74, 47–97 (2002).
- Ljung, L. System identification. (Prentice hall PTR, 1999)

#### Course policy

**Google drive**for sharing presentations and other material**Nordling Lab – ModelSelection**web page**NCKU curriculum**catalogue

Licensed under CC BY 4.0, except logos and material from published articles. |