Due the limitations of both simulation and experiment, an ultimate understanding of protein folding will likely come from a coupled approach of detailed simulations extensively validated and tested by experiment. However, developing simulation methodology which can quantitatively connect with experimental kinetics still remains a great theoretical challenge, due to the long timescales involved and the difficulties and complexities of detailed, atomistic models. Finally, with the ability of simulation to quantitatively predict experimental kinetics for protein folding in solution, new challenges emerge: understanding folding in biologically relevant contexts, such as natively unstructured proteins and folding in confined spaces. Here, we propose new, second generation distributed computing methods to tackle these challenges and the application of these methods to questions related to how proteins self-assemble in solution, as well as in the biologically relevant contexts of natively unstructured proteins and proteins folding in confined spaces as a model for important biological contexts, such as ribosomes, chaperones, and the cytosol. While protein folding has itself been studied computationally for many years, our work differs from other approaches in (1) its use of novel distributed computing methods for simulating long time scale kinetics using detailed, fully atomistic, explicit solvent models and (2) the application of these detailed models to address questions of folding in the biological contexts of natively unstructured proteins and confinement. A quantitative comparison to experiment is critical for both the testing and greater impact of our computational methods and such experimental collaborations are proposed via a series of collaborations. Finally, the proposed work should have an impact on our basic understanding of several protein- related diseases, such as cancers involved in mutations of the tetramerization and activation domains of p53. Also, by understanding the nature of folding in confined spaces, one would gain insight into the nature of protein folding in vivo, which would be an important next step in our understanding of protein folding and its connection to biology and biomedical questions. ? ? ?
| Feinberg, Evan N; Sur, Debnil; Wu, Zhenqin et al. (2018) PotentialNet for Molecular Property Prediction. ACS Cent Sci 4:1520-1530 |
| Sinitskiy, Anton V; Pande, Vijay S (2018) Computer Simulations Predict High Structural Heterogeneity of Functional State of NMDA Receptors. Biophys J 115:841-852 |
| Wagoner, Jason A; Pande, Vijay S (2018) Communication: Adaptive boundaries in multiscale simulations. J Chem Phys 148:141104 |
| Wu, Zhenqin; Ramsundar, Bharath; Feinberg, Evan N et al. (2018) MoleculeNet: a benchmark for molecular machine learning. Chem Sci 9:513-530 |
| Wang, Lee-Ping; McKiernan, Keri A; Gomes, Joseph et al. (2017) Building a More Predictive Protein Force Field: A Systematic and Reproducible Route to AMBER-FB15. J Phys Chem B 121:4023-4039 |
| Ramsundar, Bharath; Liu, Bowen; Wu, Zhenqin et al. (2017) Is Multitask Deep Learning Practical for Pharma? J Chem Inf Model 57:2068-2076 |
| Liu, Bowen; Ramsundar, Bharath; Kawthekar, Prasad et al. (2017) Retrosynthetic Reaction Prediction Using Neural Sequence-to-Sequence Models. ACS Cent Sci 3:1103-1113 |
| Harrigan, Matthew P; Sultan, Mohammad M; Hernández, Carlos X et al. (2017) MSMBuilder: Statistical Models for Biomolecular Dynamics. Biophys J 112:10-15 |
| McKiernan, Keri A; Husic, Brooke E; Pande, Vijay S (2017) Modeling the mechanism of CLN025 beta-hairpin formation. J Chem Phys 147:104107 |
| McGibbon, Robert T; Husic, Brooke E; Pande, Vijay S (2017) Identification of simple reaction coordinates from complex dynamics. J Chem Phys 146:044109 |
Showing the most recent 10 out of 126 publications
