ADVANCED QUANTITATIVE IMAGING FOR DNA PHYSICAL MAPPING

点击翻译按钮获取中文摘要

The broad goal of this multi-disciplinary project is to advance genetics research in large-scale DNA mapping by developing a computer-based system for: 1) expediting the physical mapping of large regions of DNA on the scale of individual yeast chromosomes; and 2) improving large-scale human linkage mapping. To be developed and fields is an advanced image- acquisition and processing system for extracting information from fractionated mixtures of nucleic-acid molecules separated by slab-gel electrophoresis. Specific aims are to: 1) develop a system for direct (non-photographic) imaging of gels stained with fluorescent DNA-binding dyes; 2) extend the system to accommodate autoradio-grams used in RFLP mapping; 3) develop quantitative-imaging algorithms for automated, high-resolution detection and intensity estimation of gel bands for fragment-size determination from band mobilities and for data-integrity evaluation; 4) apply methods for geometric-distortion compensation of gel images to allow direct analytic comparisons of images obtained by different methods and from different laboratories; 5) develop an interactive software system for rapid and efficient expert interpretation and analysis of digitized gel images in a state-of-the-art workstation environment; and 6) evaluate the system's effectiveness in meeting the needs of the genetic research. Methods include using an advanced CCD imaging device to avoid photographic limitations; applying locally developed image-estimation algorithms to improve quantitative analysis of DNA digested by restriction enzymes; using "warping" algorithms to compensate for electrophoretic aberrancies; performing receiver-operating-characteristic (REC) and evaluation-database studies to assess detection-algorithm performance; providing image-processing utilities for enhancing gel-image features; and aiding interactive interpretation of gel images by constructing a user interface based on accepted windowing and networking standards to ensure software portability and ease of future exportation. This three-way collaboration involves the Department of Genetics, the Department of Computer Science (CS), and the Biomedical Computer Laboratory, all at Washington University. The independently funded CS components is developing a DNA analysis system for maintaining and analyzing large amounts of both DNA-fragment-size data and gel-image data to be generated using the system here proposed.