The Human Genome Project is providing massive amounts of genetic information that should revolutionize the understanding and diagnosis of inherited diseases. In particular, it is hoped that the detection of single nucleotide polymorphism (SNPs) in gene coding and regulatory regions will lead to a greater comprehension of the genetic contribution to risk for cancer, to the elucidation of the genetic factors that affect treatment and prevention, and to the design of powerful new drugs. Achieving these goals will require substantial effort in several areas, including detection or discovery of SNPs and development of high throughput methods for characterizing and comparing nucleotide variations in specific regions of the genome. In addition, recent observations suggest that the haplotype of a subject - the specific alleles associated with each chromosome homologue - is a critical element in SNP mapping, although current methods for determining haplotypes have significant limitations that have prevented their use in large-scale genetic screening. To address this substantial technology gap and provide critical information for molecular level cancer research and medicine, this project will develop a novel technology, which is based upon direct molecular scale imaging of DNA using carbon nanotube atomic force microscopy probes, for the detection and characterization of nucleotide variations, and in particular, direct haplotype determination. In the R21 phase of the project, we will develop the materials science and chemistry required to make nanotube probe tips that provide reproducible detection capability, and to integrate methods from biology, chemistry and engineering for preparation and deposition of DNA samples suitable for reproducible analysis. In the subsequent R33 phase of the project, we will be extend significantly the throughput of our novel technology by developing (i) an integrated system capable of automated deposition of sample arrays, array imaging and image analysis, (ii) system and software to increase significantly sample detection throughput, and (iii) multiprobe arrays for ultrahigh throughput parallel sample imaging.