The National Center for Supercomputing Applications has selected nine U. of I. researchers to receive NCSA fellowships with research support that will enable them to pursue collaborative projects with the researchers and computer technology experts at NCSA. Fellows play an important role at NCSA as the center implements its "big research, big data, big compute" strategy to develop long-term research activities addressing complex grand challenge-scale problems across science, engineering, humanities and the arts.
"I hope that these fellowships will launch new and lasting partnerships and connections between NCSA and other University of Illinois research leaders," said NCSA director Ed Seidel.
Gabrielle Allen, NSCA associate director, directs NCSA's fellowship program.
The 2014-2015 NCSA fellows:
Scott Althaus, a professor of political science and of communication, "Building a Non-Consumptive Global News Observatory for Data Science Research." A relatively new approach to research on copyrighted materials offers a way to provide full-text access to data science researchers without violating copyright or requiring database distributors to forfeit proprietary control over their investments. Known as "non-consumptive research," this new paradigm allows researchers to run algorithms on copyrighted full-text holdings without allowing them to see or copy those holdings. This project will prototype a non-consumptive research platform for text mining the Cline Center's full-text news holdings.
Narayana Aluru, a professor of mechanical science and engineering, "Nanopore Genome." Nanopore technology has gained significant prominence because of its ability to provide breakthrough solutions to societal problems. The goal of this project is to create a data resource of soft matter properties in nanopores. Specifically, structure, dynamics and transport properties of soft matter in a variety of nanoporous materials will be computed and made available to the international community as part of the data resource. Blue Waters enables rapid calculation of soft matter properties in nanopore materials, and the National Data Service will provide the necessary cyberinfrastructure and the user interfaces for data storage and access.
Rosa M. Espinosa-Marzal, a professor of civil and environmental engineering, "Critical Molecular Interactions Underlying Biomineralization." Biomineralization, a process by which organisms precipitate inorganic minerals, is a widespread phenomenon in nature. Biominerals have superior properties, such as high toughness and strength, and self-healing ability. This project will provide understanding of the molecular interactions responsible for mineral formation and self-healing properties by directly exploring them through ab initio quantum chemistry methods. Molecular-scale simulations are key for the required level of understanding. The long-term goal is to establish design principles for sustainable self-healing biomimetic materials.
Matthew Hudson, a professor of bioinformatics and of genomics, "Development and Deployment of a Sequencing and Bioinformatics Pipeline for Rapid Detection of Structural Variations in Crop Genomes." Crop plants have genomes as large and complex as the human genome - in some cases larger and more complex. However, almost all bioinformatics tools currently available are written for the human genome. The project will develop a variant calling pipeline that uses high-performance computing resources to rapidly and accurately call all the major types of variants responsible for valuable traits in crops, based on high-throughput sequence data. A cross-platform, open-source package will be published at the end of the project for the crop science community.
Iwona Jasiuk, a professor of mechanical science and engineering, "Multiscale Modeling of Bone Fracture and Strength." This project will use a novel multiscale computational approach to predict fracture and strength of normal versus osteoporotic bone. This project will create a multiscale computational model of bone fracture and strength. The complex hierarchical structure of bone will be analyzed in an "ascending" order by considering finest details of substructures, which serve as building blocks of bone. This will be the first experimentally-based multiscale model of bone fracture and strength, and it should have high impact on clinical assessment of bone in health and disease.
David LeBauer, a research scientist in the Institute for Genomic Biology, "Coupling Meteorology, Plant Biology and Economic Engineering Models Within a CyberGIS Framework." This project will integrate existing models and workflow tools used to simulate weather, bioenergy production and supply chain optimization within a CyberGIS platform to support scientific investigations into the production and distribution of biofuel feedstocks. This will facilitate cross-domain inquiry and collaboration, increase the scope and precision of scientific inference, and improve the accessibility of complex computational workflows to scientists and decision makers.
Taras Pogorelov, a senior research scientist in the School of Chemical Sciences, and a research professor in the department of chemistry and the Beckman Institute for Advanced Science and Technology, "An Open Environment for Automation of Molecular Dynamics Simulations of Membrane-Active Host Defense Peptides." Molecular dynamics methods changed the way chemists and biologists approach the computational material design problems. The complexity of connecting the macroscopic phenomena with atomistic processes makes conducting computational biophysics simulations on high-performance computing (HPC) resources labor-intensive. This forces researchers to spend valuable time on conducting the simulations, rather than focusing on the science problem. The goal of this project is to develop the architecture design for an extensible software package, the Computational Discovery Desktop, that will provide a customizable human-HPC interface to conduct computational projects in a manner requiring minimal user intervention, encouraging the reuse of plug-in components, and providing management of computational projects.
John Toenjes, a professor and the music director for the department of dance, "Laboratory for Audience Interactive Technologies." Because of the connectivity, programmability and ubiquity of modern mobile devices, the theatrical world is on the cusp of a sea change in the relationship between the audience and the performers; among the performers themselves; and between performers, audience and the theatrical stage. This project will investigate the expressive, informational and integrative possibilities presented by personal digital devices and technologies such as augmented reality, Web and phone apps, and social media/Internet connectivity. LAIT will serve as an idea vetting house, an R&D laboratory, and a beta rollout service for artistic and informational apps for the art and entertainment, installation, education and information industries.
Daniel Work, a professor of civil and environmental engineering, "Improving the Efficiency of Taxi Systems Through Real-time Seek Time Prediction." The goal of this project is to assess if the seek time for a taxi is predictable, and if so, to also learn the predictors. The seek time for a taxi is the time between when the previous passenger was dropped off and when the next passenger is picked up. High driver income correlates with low seek times between trips. This work will examine a dataset of more than 700 million taxi trips in New York City to predict seek times, ultimately enabling predictive taxi information services to improve system efficiency.