Showing 2 Results
Showing 1-2 of 2
Publications
Authors: Ingólfsson, Helgi I., Neale, Chris, Carpenter, Timothy S., Shrestha, Rebika, López, Cesar A., Tran, Timothy H., Oppelstrup, Tomas, Bhatia, Harsh, Stanton, Liam G., Zhang, Xiaohua, Sundram, Shiv, Natale, Francesco Di, Agarwal, Animesh, Dharuman, Gautham, Schumacher, Sara I. L. Kokkila, Turbyville, Thomas, Gulten, Gulcin, Van, Que N., Goswami, Debanjan, Jean-Francois, Frantz, Agamasu, Constance, Chen, De, Hettige, Jeevapani J., Travers, Timothy, Sarkar, Sumantra, Surh, Michael P., Yang, Yue, Moody, Adam, Liu, Shusen, Essen, Brian C. Van, Voter, Arthur F., Ramanathan, Arvind, Hengartner, Nicolas W., Simanshu, Dhirendra K., Stephen, Andrew G., Bremer, Peer-Timo, Gnanakaran, S., Glosli, James N., Lightstone, Felice C., McCormick, Frank, Nissley, Dwight V., Streitz, Frederick H.
TITLE: Machine Learning–driven Multiscale Modeling Reveals Lipid-dependent Dynamics of RAS Signaling Proteins , National Academy of Sciences , 1 , 119 , 2022
PUBLICATION DATE: 01-04-2022
ABSTRACT: RAS is a signaling protein associated with the cell membrane that is mutated in up to 30% of human cancers. The community has proposed that RAS signaling is regulated by dynamic heterogeneity of the cell membrane. Investigating such a mechanism requires near-atomistic detail at macroscopic temporal and spatial scales, which is not possible with conventional computational or experimental techniques. The authors demonstrate here a multiscale simulation infrastructure that uses machine learning to create a scale-bridging ensemble of over 100,000 simulations of active wild-type KRAS on a complex, asymmetric membrane. Initialized and validated with experimental data (including a new structure of active wild-type KRAS), these simulations represent a substantial advance in the ability to characterize RAS-membrane biology. The authors report distinctive patterns of local lipid composition that correlate with interfacially promiscuous RAS multimerization. These lipid fingerprints are coupled to RAS dynamics, predicted to influence effector binding, and therefore may be a mechanism for regulating cell signaling cascades.
PROJECT: ADMIRRAL
Go To Publications
https://www.pnas.org/content/119/1/e2113297119
Authors: Di Natale, Francesco, Bhatia, Harsh, Carpenter, Timothy S., Neale, Chris, Kokkila-Schumacher, Sara, Oppelstrup, Tomas, Stanton, Liam, Zhang, Xiaohua, Sundram, Shiv, Scogland, Thomas R. W., Dharuman, Gautham, Surh, Michael P., Yang, Yue, Misale, Claudia, Schneidenbach, Lars, Costa, Carlos, Kim, Changhoan, D'Amora, Bruce, Gnanakaran, Sandrasegaram, Nissley, Dwight V., Streitz, Fred, Lightstone, Felice C., Bremer, Peer-Timo, Glosli, James N., Ingólfsson, Helgi I.
TITLE: A Massively Parallel Infrastructure for Adaptive Multiscale Simulations: Modeling RAS Initiation Pathway for Cancer , International Conference for High Performance Computing, Networking, Storage, and Analysis : 1–16 , 2019
PUBLICATION DATE: 11-17-2019
ABSTRACT: Computational models can define the functional dynamics of complex systems in exceptional detail. However, many modeling studies face seemingly incommensurate requirements: Gaining meaningful insights into some phenomena requires models with high resolution (microscopic) detail that must nevertheless evolve over large (macroscopic) length- and time-scales. Multiscale modeling has become increasingly important to bridge this gap. Executing complex multiscale models on current petascale computers with high levels of parallelism and heterogeneous architectures is challenging. Many distinct types of resources need to be simultaneously managed, such as graphics processing units (GPUs) and central processing units (CPUs), memory size and latencies, communication bottlenecks, and filesystem bandwidth. In addition, robustness to failure of compute nodes, network, and filesystems is critical. The authors introduced a first-of-its-kind, massively parallel Multiscale Machine-Learned Modeling Infrastructure (MuMMI), which couples a macro scale model spanning micrometer length- and millisecond time-scales with a micro scale model employing high-fidelity molecular dynamics (MD) simulations. MuMMI is a cohesive and transferable infrastructure designed for scalability and efficient execution on heterogeneous resources. A central workflow manager simultaneously allocates GPUs and CPUs while robustly handling failures in compute nodes, communication networks, and filesystems. A hierarchical scheduler controls GPU-accelerated MD simulations and in situ analysis. The authors presented the various MuMMI components, including the macro model, GPU-accelerated MD, in situ analysis of MD data, machine learning selection module, a highly scalable hierarchical scheduler, and the central workflow manager that ties these modules together. In addition, the authors presented performance data from our runs on Sierra, in which the authors validated MuMMI by investigating an experimentally intractable biological system: the dynamic interaction between RAS proteins and a plasma membrane. The authors used up to 4000 nodes of the Sierra supercomputer, concurrently utilizing over 16,000 GPUs and 176,000 CPU cores, and running up to 36,000 different tasks. This multiscale simulation includes about 120,000 MD simulations aggregating over 200 milliseconds, which is orders of magnitude greater than comparable studies.
PROJECT: ADMIRRAL
Go To Publications
https://doi.org/10.1145/3295500.3356197
Download XLSX