HIGH PERFORMANCE BIOMEDICAL COMPUTING

点击翻译按钮获取中文摘要

The goals of the high performance biomedical computing program are to identify and solve those computational problems in biomedicine that can benefit from high performance hardware, modern software engineering principles, and efficient algorithms. This effort includes providing high performance parallel computer systems for the NIH staff and developing parallel algorithms for biomedical applications. CBEL is developing algorithms for a number of biomedical applications that can benefit from computational speedup including image processing of electron micrographs, protein and nucleic acid sequence analysis, nuclear magnetic resonance spectroscopy, x-ray crystallography, protein folding prediction, quantum chemical methods, molecular dynamics simulations, human genetic linkage analysis, medical imaging, and radiation treatment planning. Developing teams for each application area include computer engineers and scientists from CBEL who design and implement the required parallel algorithms and methods, and biomedical scientists who provide the necessary application knowledge and become users of the developed software. The ultimate goal is to have high performance parallel computing facilitate the science that is done at NIH. While developing these computationally demanding applications, CBEL is investigating the following high performance computing issues: partitioning a problem into many parts that can be independently executed on different processors, designing algorithms so that delays of interprocessor communication can be kept to a small fraction of the computation time, designing the parts so that the computing load can be distributed evenly over the available processors or dynamically balanced, designing algorithms so that the number of processors is a parameter and the algorithms can be configured dynamically for the available machine, developing tools and environments for producing portable parallel programs and monitoring system performance, and proving that a parallel algorithm on a given machine meets its specifications. The work of CBEL staff has contributed to a number of findings of biomedical significance in the past year. Working with NIAMS collaborators, progress was made on determining the three-dimensional location of the major capsid proteins of the herpes simplex virus (type 1). NIDDK has used parallel computing methods to improve their procedure for determining the structure of the protein Calmodulin from NMR spectra data. Another group of scientists from NIDDK simulated the kinetics of nitric oxide rebinding to myoglobin following photodissociation on the CBEL parallel computer. NIMH investigators used parallel image registration techniques developed by CBEL staff to study the progression of Alzheimer's disease from PET images of the brain. High performance computing has allowed NEI researchers to determine the onset time, the rate of information encoding, and the total amount of information encoded by the neuronal responses to different parameters of a visual stimulus in primates.