Browsing by Author "Wolynes, Peter"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    cell motility and machine learning
    (2020-03-11) Yu, Guangyuan; Wolynes, Peter; Levine, Herbert
    Cell migration is a necessary function in organisms. It is relevant to wound healing, immune reaction, cancer expansion. One example is durotaxis: Cells exhibit qualitatively different behaviors on substrates with different rigidities. The fact that cells are more polarized on the stiffer substrate motivates us to construct a two-dimensional cell with the distribution of focal adhesions dependent on substrate rigidities. This distribution affects the forces exerted by the cell and thereby determines its motion. Our model reproduces the experimental observation that the persistence time is higher on the stiffer substrate. This stiffness-dependent persistence will lead to durotaxis, the preference in moving towards stiffer substrates. We derive and validate a two-dimensional corresponding Fokker-Planck equation associated with our model. Another example is the chemotaxis: Leukotriene B4 is secreted as exosomes by neutrophils, serving as a secondary gradient to amplify the chemical attraction of primary chemoattractants. We introduce a model to compare the enhancement effect between directly releasing Leukotriene B4 and generating as exosomes. We attribute this advantage to the longer lifetime of exosomes which leads to a larger range of attraction. The third example is in the immune system: The infiltrations of T are different in patients, which could be a tool for the prognosis. High CD8+ T cell counts (both overall and inside cancer-cell islands) is associated with better patient outcome. However, a cut-off of the T-cell count has to be selected manually to separate groups of patients. In this work, we propose a method to classify the small patch of triple-negative breast cancer (TNBC) tumor and use the overall percentage of good patches as a marker to predict the prognosis, which is an automatic method of prognosis and could also be used for other cancers. The result shows that the machine learns the importance of cell count and cell infiltration and use the combination as an indicator for prognosis. We also applied the machine learning method on sperm classification and quantum many body problem. In sperm classification, we could get around 70\% accuracy with balanced good and poor example. In quantum many body problem, we proposed a deep neural network to calculate the ground-state energies and it shows excellent agreement with the Bethe-Ansatz exact solution. Furthermore, we also calculate the loop correlation function using the wave function obtained.
  • Thumbnail Image
    Item
    Modeling protein structural ensembles using AWSEM-Suite
    (2023-04-11) Jin, Shikai; Tao, Jane; Wolynes, Peter
    Proteins are the driving force behind most cellular processes. Traditional methods for determining protein structure are limited to probing only a few static structures of a given protein. However, molecular dynamics simulation presently allows for the determination of the dynamics of a protein. In this thesis, I introduce AWSEM-Suite, a coarse-grained force field that has recently shown good performance in protein structure prediction experiments. This force field is based on physical principles and neural network-based machine learning. I describe the composition of the force field, as well as two new energy terms: the template-based and coevolutionary-guided terms. After discussing the problem of protein structure prediction, I showcase how we built an online server for AWSEM-Suite, describing the path between the input and output. Additionally, I discuss other techniques for structural analysis, including frustration protein structure refinement, a physics-based method that complements current methods. In the last part of this thesis, I introduce two applications: solving phase problems in X-ray crystallography and exploring the translocation mechanism of bacteriophage T7 helicase gp4 using AWSEM-Suite. Our results show that AWSEM-Suite provides better results than the state-of-the-art program I-TASSER-MR in finding phase information for a given protein. Furthermore, AWSEM-Suite can successfully predict the key loops that interact with ssDNA during gp4 translocation. We explore the intermediate structures of action, highlighting the possible mechanism of the role of electrostatic effects exerted by the binding and release of ATP molecules. In these chapters, we explore several related questions concerning protein structure prediction, protein structure refinement, and specific biological questions using AWSEM-Suite. In summary, our studies demonstrate that AWSEM-Suite is a powerful technology for exploring protein dynamics and predicting protein structure.
  • Thumbnail Image
    Item
    Pseudogene Energy Landscapes: A Frustrating Case of Neutral Evolution?
    (2020-10-07) Jaafari, Hana; Wolynes, Peter
    Functional proteins are optimized by evolution over the course of millions of years to quickly fold into their native three-dimensional structures. Evolution exerts a strong selection pressure on protein sequences to maintain foldability and stability, resulting in minimally frustrated folding energy landscapes. Pseudogenes, genetic elements homologous to coding genes, are evolutionary relics of the genome experiencing little or no selection pressure. Pseudogenes are ideal candidates to examine the energy landscapes of devolving genetic elements. This thesis project is the first to quantify the “evolutionary” energies, measured with Direct Coupling Analysis (DCA), of pseudogenes across multiple protein families. The DCA energies of pseudogenes, their parent proteins, and other proteins within each family were examined, and the results of these studies suggest that pseudogenes become less well optimized from an evolutionary standpoint over time. Indeed, analyses of the DCA energies of mutants generated in silico indicated that pseudogenes devolve just as rapidly as completely randomly mutated parent genes.