Understanding the Markov State Model and Its Application to the PPAR-gamma System
(2025) KFKM01 20251Biophysical Chemistry
- Abstract
- Peroxisome proliferator-activated receptor gamma (PPAR-gamma) is a key regulator of metabolism, but how its structural dynamics control allosteric signaling remains unclear. To investigate this, we performed molecular dynamics simulations of PPAR-gamma and applied time-lagged independent component analysis to reduce the dimensionality of the data. Microstates were identified using K-means clustering and then coarse-grained into macrostates using Perron-cluster cluster analysis. We further computed mean first-passage times to quantify the transition kinetics between these states. A three-dimensional projection of the dynamics enabled the identification of four major macrostates. The S4 state showed the highest population, accounting for 80%... (More)
- Peroxisome proliferator-activated receptor gamma (PPAR-gamma) is a key regulator of metabolism, but how its structural dynamics control allosteric signaling remains unclear. To investigate this, we performed molecular dynamics simulations of PPAR-gamma and applied time-lagged independent component analysis to reduce the dimensionality of the data. Microstates were identified using K-means clustering and then coarse-grained into macrostates using Perron-cluster cluster analysis. We further computed mean first-passage times to quantify the transition kinetics between these states. A three-dimensional projection of the dynamics enabled the identification of four major macrostates. The S4 state showed the highest population, accounting for 80% of the total, and possessed the lowest free energy. Analysis of the transition kinetics indicated that the transition into the S1 state was the rate-limiting step. Although the resulting Markov State Model captures key features of PPAR-gamma conformational dynamics, the model is likely influenced by sensitivity to parameter choices and simplifications introduced during state decomposition. These results should therefore be interpreted as a preliminary framework for understanding allosteric signaling in PPAR-gamma, rather than a definitive description. (Less)
- Popular Abstract
- Mapping a "Trap Door" in a Master Switch for Metabolism
At the heart of common diseases like type 2 diabetes is a tiny molecular "master switch" called PPAR-gamma. To design better drugs, we need to understand how this switch moves, but the crucial journey between its different shapes has been mostly invisible. Using supercomputers, we created a detailed map of this protein’s movements, which suggests it has a natural "trap door": a specific shape that it falls into easily but escapes from very slowly. Proteins are the tiny machines running our bodies, and PPAR-gamma is one of the most important. It controls genes related to how we handle fat and sugar. While we have static "photographs" of its key functional shapes from experiments, we... (More) - Mapping a "Trap Door" in a Master Switch for Metabolism
At the heart of common diseases like type 2 diabetes is a tiny molecular "master switch" called PPAR-gamma. To design better drugs, we need to understand how this switch moves, but the crucial journey between its different shapes has been mostly invisible. Using supercomputers, we created a detailed map of this protein’s movements, which suggests it has a natural "trap door": a specific shape that it falls into easily but escapes from very slowly. Proteins are the tiny machines running our bodies, and PPAR-gamma is one of the most important. It controls genes related to how we handle fat and sugar. While we have static "photographs" of its key functional shapes from experiments, we don’t fully understand the dynamic pathways it takes to switch between them.
To see this journey, we created a kind of molecular movie using massive computer simulations. This allowed us to watch the PPAR-gamma protein jiggle, twist, and change its shape billions of times. The problem is that this "movie" is far too long and complex for any person to watch. To make sense of it, we used a special analysis method called a Markov State Model, which acts like a map-making tool. It automatically sorts through all the chaos to produce a simplified "subway map" of the protein’s favorite shapes and the main routes it takes to travel between them.
Our map suggests a fascinating mechanism. It indicates that PPAR-gamma, even when empty, prefers four main shapes. One of them, the "trap door" state, appears to be its most stable, lowest-energy conformation. This is surprising because this shape structurally resembles the form the protein takes when a drug is actively turning it off. Even more surprising, we found that when we started our simulation from the known "empty" crystal structure, the protein didn’t relax into this stable off state. Instead, it often became stuck in a much less stable, high-energy shape, as if it were primed for activation. It is important to be honest about the limitations of such a computer model. The results can be sensitive to the parameters chosen, and we were not able to perfectly resolve the very slowest motions. Therefore, these findings should be viewed as a strong scientific hypothesis, not a final truth.
So, how can this work be used? This model provides a new, detailed roadmap for drug development. By revealing the protein’s preferred shapes and the routes between them, our work helps lay the foundation for new therapeutic strategies. Future studies could investigate how drugs might alter this energy landscape, potentially leading to medicines that work by stabilizing specific conformations, such as the low-energy "trap door" state we identified. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9201461
- author
- Lu, Yao LU
- supervisor
-
- Pär Söderhjelm LU
- Mikael Akke LU
- organization
- course
- KFKM01 20251
- year
- 2025
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- Biophysical Chemistry, Molecular Dynamics, Markov State Model, Protein Dynamics, Allosteric Signaling, PPAR-gamma
- language
- English
- id
- 9201461
- date added to LUP
- 2025-06-18 14:36:44
- date last changed
- 2025-06-18 14:36:44
@misc{9201461,
abstract = {{Peroxisome proliferator-activated receptor gamma (PPAR-gamma) is a key regulator of metabolism, but how its structural dynamics control allosteric signaling remains unclear. To investigate this, we performed molecular dynamics simulations of PPAR-gamma and applied time-lagged independent component analysis to reduce the dimensionality of the data. Microstates were identified using K-means clustering and then coarse-grained into macrostates using Perron-cluster cluster analysis. We further computed mean first-passage times to quantify the transition kinetics between these states. A three-dimensional projection of the dynamics enabled the identification of four major macrostates. The S4 state showed the highest population, accounting for 80% of the total, and possessed the lowest free energy. Analysis of the transition kinetics indicated that the transition into the S1 state was the rate-limiting step. Although the resulting Markov State Model captures key features of PPAR-gamma conformational dynamics, the model is likely influenced by sensitivity to parameter choices and simplifications introduced during state decomposition. These results should therefore be interpreted as a preliminary framework for understanding allosteric signaling in PPAR-gamma, rather than a definitive description.}},
author = {{Lu, Yao}},
language = {{eng}},
note = {{Student Paper}},
title = {{Understanding the Markov State Model and Its Application to the PPAR-gamma System}},
year = {{2025}},
}