You are here
Algorithms developed by Professor Vipin Kumar (Computer Science and Engineering, MSI Fellow) and his team can be used to detect changes in forest cover due to fires, logging, and other events. Deforestation is an area of concern to scientists studying climate change. Professor Kumar’s software is being used by the Planetary Skin Institute as part of their effort to integrate data from sensors monitoring the earth. The graph shows changes in vegetation over time detected by the algorithm using satellite imagery. A story about the Kumar group’s work appeared in the Fall 2008 MSI Research Bulletin. The University News Service story about this work can be found here.
Associate Professors Hiroshi Matsuo (MSI Associate Fellow) and Reuben Harris of the Department of Biochemistry, Molecular Biology, and Biophysics lead a team of researchers studying ways of fighting HIV, the virus that causes AIDS. Specifically, they are studying a protein called APOBEC3G, or A3G, which can alter the HIV genome by deaminating cytosines to uracils. (Cytosine is one of the bases in DNA; uracil is a base in RNA.) The group’s work was featured in the prestigious journal Nature in 2008, and Professor Harris was recently awarded a $100,000 Grand Challenges Explorations grant from the Bill and Melinda Gates foundation to continue this important research. The graphic shown is an example of a computer-generated image of a model used by the team to graphically depict molecular structures, in this case a superimposition of ten NMR images (modified from Figure 2 in Chen et al., Nature, 452, 116-119 (2008)).
Professors Lynda Ellis, Department of Laboratory Medicine and Pathology, and Larry Wackett, Department of Biochemistry, Molecular Biology and Biophysics, both MSI Associate Fellows, have co-directed the University of Minnesota Biocatalysis/Biodegradation Database (UM-BBD) since 1995. The UM-BBD, hosted at MSI, contains information on microbial biodegradation pathways, primarily for environmental pollutants. The UM-BBD Pathway Prediction System (UM-PPS) predicts microbial metabolic pathways, to help scientists determine how microbes may clean up an environment contaminated with new chemical compounds. The image at left is part of a visualization of a UM-PPS prediction for the microbial degradation of 2-bromo-1,1-dichloroethane.
It’s not easy being a plant. There are any number of pathogens just waiting to attack you, and, being a plant, you can’t run away. Fortunately, plants can defend themselves, and the research group of Associate Professor Fumiaki Katagiri, Department of Plant Biology and MSI Associate Fellow, is studying how. Their cross-disciplinary research, which involves molecular biology, biochemistry, genetics, reverse genetics, genomics, expression profiling, proteomics, structural biology, and computational biology, investigates how plants recognize the molecular signals of pathogen attack and how they then coordinate their defense responses. This work may lead to disease-control methods safer for humans and for the environment. The image at left is a method of visualizing relationships among mutant plants.
The use of massively parallel computing on powerful machines has been a great benefit to many scientific researchers. Among these are the astrophysicists, who deal with enormous datasets as they study cosmic phenomena. Professor Tom Jones, Department of Astronomy and Interim Director of the Supercomputing Institute, and his research group have successfully developed a fully parallel code that scales to several thousand cores. Using this code, the group is able to create simulations of galaxy jets. The simulation shown is a 2000x2000x500 mesh simulation of a "narrow angle tail” galactic jet, which is due to a relatively weak jet in a strong wind; the picture shows the jet’s vorticity in log scale. This computation was performed on a cluster running Microsoft HPC Server 2008.