Souvik-SNH / qMSM_tutorial

A Tutorial for quasi Markov State Model(qMSM) developed by Huang Group, Dept of Chemistry at UW-Madison

A Step-by-step guide for building quasi Markov State Model to Study Functional Conformational Changes of Biological Macromolecules

Version 0.1, (c) Huang Group, Department of Chemistry, University of Wisconsin-Madison

This is a tutorial for constructing quasi-Markov State model(qMSM), a new kinetic modelling framework that encodes non-Markovian dynamics in time-dependent memory kernels.

The following content is accompanying the book chapter:

"A Step-by-step Guide on How to Construct quasi-Markov State Models to Study Functional Conformational Changes of Biological Macromolecules", Chapter 10, “A Practical Guide to Recent Advances in Multi-scale Modeling and Simulation for Biomolecules”, Edited by Wang, Y. and Zhou, R., AIP Publishing (2023).

(Freely available at ChemRxiv: https://chemrxiv.org/engage/chemrxiv/article-details/620b1bff0c0bf09733e9cde8)

Content

There are 8 Jupyter notebooks that depict various stages of qMSM construction

1. Featurization Link to notebook
2. Feature selection with Spectral oASIS Link to notebook
3. Dimensionality reduction with tlCA Link to notebook
4. Clustering Link to notebook
5. MSM hyperparameter selection with GMRQ Link to notebook
6. Microstate MSM and Lumping Link to notebook
7. quasi-Markov State Model Link to notebook
8. MFPT calculation and macrostate sampling Link to notebook

For the MD trajectories, you may download them at our dispository on Open Science Framework: https://osf.io/wu2s6/?view_only=c7c5fef31563409babb403669a864572

Installation

We will use MSMbuilder 3.8.0 and PyEMMA in our tutorial. For installing MSMbuilder with Anaconda, you can use the following script:

conda env create -n msmbuilder -f environment.yml

For PyEMMA installation, you may refer to http://pyemma.org.

Authors

• Prof. Xuhui Huang - Project leader - xhuang
• Andrew Kai-hei Yik - Author
• Ilona Christy Unarta - Author
• Yunrui Qiu - Author
• Siqin Cao - Author

License

This tutorial is licensed with Creative Commons Attribution 4.0 International License. The library for Quasi-MSM construction is licensed with Apache License, Version 2.0.

Note for readers and people curious about qMSM

You may contact Dr Siqin Cao [e-mail]/ Qiu Yunrui [e-mail] in case you have any specific questions on the qMSM python package or model construction using the qMSM framework.