News

IACAT fellowships support diverse collaborative projects

Published Date: Jul 09, 2012

Thirteen researchers from the University of Illinois at Urbana-Champaign have been selected to receive one-year fellowships that will enable them to pursue collaborative projects with the researchers and computer technology experts at the National Center for Supercomputing Applications.

The 2012-2013 IACAT fellows are:

  • Weng Cho Chew (Electrical and Computer Engineering), working with Narayan Aluru (Mechanical Science and Engineering) and NCSA’s John Towns, aims to improve two algorithms used to model Casimir forces, which arise between objects at very small length scales and can affect performance of micro- and nanoelectromechanical systems. The team’s goal is to increase by orders of magnitude the size of the problems that can be solved.

  • Robert B. Haber (Mechanical Science and Engineering) will work with NCSA’s Volodymyr Kindratenko to explore the use of graphics-processing units (GPUs) to accelerate the performance of a solver used to model hyperbolic systems.

  • Xiuling Li (Electrical and Computer Engineering), in collaboration with K. Jimmy Hsia (Mechanical Science and Engineering) and Seid Koric (NCSA), will model the rolling process of strain-induced rolled-up micro- and nanostructures. Understanding of this process could provide guidance for the design and assembly of nanostructures into systems that can adapt to environmental changes in heat, optical radiation, and mechanical actuation.

  • Robert Markley (English) will team with Kenton McHenry, leader of NCSA’s Image, Spatial, Data Analysis Group, and Michael Simeone of the Institute for Computing in the Humanities, Arts, and Social Science (I-CHASS) to build on a previous computational analysis of 17th and 18th-century maps of the Great Lakes region, adding approximately 400 maps to the data set to test a hypothesis that the maps encode short-term meteorological and long-term climatological data in the form of variations in cartographic depictions of coastlines, islands, and water passages among the lakes.

  • Surangi W. Punyasena (Plant Biology) will work with David Tcheng (Illinois Informatics Institute) and NCSA’s Michael Welge. The team plans to use advanced machine learning algorithms to develop a high-throughput automated system to count pollen and spore samples from fossils. This will make it feasible to create large data sets, which is impractical when counts are taken by individual highly trained experts using light microscopes.

  • Ted Underwood (English) will work with Loretta Auvil and Boris Capitanu from the Illinois Informatics Institute and NCSA’s Mark Straka to scale up an algorithm for topic modeling—a technique for mapping the topics that could have produced a given collection of documents. The research team previously established that this method can reveal significant patterns in literary history.

  • Robert Brunner (Astronomy) will work with NCSA’s Don Petravick to develop a new technique for distinguishing stars and galaxies in large, photometric datasets. Data from the Sloan Digital Sky Survey will be used to develop the new technique, which then will be applied to the upcoming Dark Energy Survey, for which NCSA will serve as the data archive.

  • Prashant Jain (Chemistry) will work with NCSA’s Sudhakar Pamidighantam to develop a platform for computational characterization and prediction of novel nano-optic phenomena. This platform will provide key design guidelines for engineering of optical resonances and electromagnetic fields using metal/semiconductor nanostructures with utility in enhanced light harvesting, ultrasensitive sensing, and induction of non-natural light-matter interactions in materials and molecules.

  • Logan Liu (Electrical and Computer Engineering) and NCSA’s Jong Lee plan to develop handheld water sensors that use mobile smartphones’ signal acquisition, processing, and transmission ability. Smartphone users will be able to measure soluble chemicals in water samples, and the quantitative analysis results (along with geospatial locations and temporal tags) will be pushed to supercomputing resources at NCSA. When hundreds and even thousands of “citizen environmental inspectors” are equipped with the sensors, regional water quality maps will be created and made publicly available.

  • Juan Loor (Animal Sciences) will work with the data analytics team led by NCSA’s Michael Welge to develop a visual, interactive tool for studying and exploring bovine transcriptome data from previous studies that dealt with different nutritional interventions during late-pregnancy through early lactation. The data from liver, adipose tissue, and mammary gland tissue will provide a systems understanding of the underlying clusters and networks of genes that coordinate the functional adaptations of these tissues to the onset of lactation. The dynamics of the transcriptome will be assembled into a more holistic model by including whole-animal level data from the same animals, e.g. milk production rate, changes in body weight, and changes in blood metabolites and hormones.

  • Momcilo Markus (Illinois State Water Survey)—in collaboration with Kenton McHenry, leader of NCSA’s Image, Spatial, Data Analysis group, and atmospheric science researcher Don Wuebbles—will use computational modeling to predict the flood peaks in northeastern Illinois for selected future climate scenarios. To do this, the team will modify the Variable Infiltration Capacity (VIC) hydrologic model for applicability to urban areas and will calibrate the model for present climate conditions.

  • Luke Olson (Computer Science) will work with Jeremy Enos, Volodymyr Kindratenko, and Mike Showerman at NCSA to research more energy-efficient solvers for sparse matrix computations, which play a major role in a wide variety of scientific simulations and represent a significant amount of both total simulation time and power consumption.

  • Rachel Whitaker (Microbiology) and collaborators at the Georgia Institute of Technology have developed a multi-scale model of host-viral evolution that integrates molecular, ecological, and evolutionary parameters. Working with NCSA’s Mark Straka the team will port the code to high-performance computing resources and perform large-scale simulations of the co-evolution of hosts and pathogens.


For more information about the IACAT Fellows program, see http://iacat.illinois.edu/faculty_fellows.