Last year, the X27A beamline at the National Synchrotron Light Source (NSLS) became dedicated solely to X-Ray Computed Microtomography (XCMT). This is a third-generation instrument capable of producing tomographic volumes of 1-2 micron resolution over a 2-3mm field of view. Recent enhancements will be discussed. These have focussed on two issues: the desire for real-time data acquisition and processing and the need for highly monochromatic beam (.1 % energy bandpass). The latter will permit k-edge subtraction studies and will provide improved image contrast from below the Cr (6 keV) up to the Cs (36 keV) k-edge. A range of applications that benefit from these improvements will be discussed as well. These two goals are somewhat counterproductive, however; higher monochromaticity yields a lower flux forcing longer data acquisition times. To balance the two, a more efficient scintillator for X-ray conversion is being developed. Some testing of a prototype scintillator has been performed; preliminary results will be presented here. In the meantime, data reconstruction times have been reduced, and the entire tomographic acquisition, reconstruction and volume rendering process streamlined to make efficient use of synchrotron beam time. A Fast Filtered Back Transform (FFBT) reconstruction program recently developed helped to reduce the time to reconstruct a volume of 150 x 150 x 250 pixels 3 (over 5 million voxels) from the raw camera data to 1.5 minutes on a dual R10,000 CPU. With these improvements, one can now obtain a "quick look" of a small tomographic volume (~l0 6 voxels) in just over 15 minutes from the start of data acquisition.

The problem of determining an expression whose derivative is a specified expression when that expression involves unspecified functions is analyzed. An algorithm is presented which reduces this problems to that of integration of specified functions.

A number of essential biological functions are controlled by proteins that bind to specific sequences in genomic DNA. In this article we present a simplified model for analyzing DNA‐protein interactions mediated exclusively by hydrogen bonds. Based on this model, an optimized algorithm for geometric pattern recognition was developed. The large number of local energy minima are efficiently screened by using a geometric approach to pattern matching based on a square‐well potential. The second part of the algorithm represents a closed form solution for minimization based on a quadratic potential. A Monte Carlo method applied to a modified Lennard‐Jones potential is used as a third step to rank DNA sequences in terms of pattern matching. Using protein structures derived from four DNA‐protein complexes with three‐dimensional coordinates established by X‐ray diffraction analysis, all possible DNA sequences to which these proteins could bind were ranked in terms of binding energies. The algorithm predicts the correct DNA sequence when at least two hydrogen bonds per base pair are involved in binding to the protein, providing a partial solution to the three‐dimensional docking problem. This study lays a framework for future refinements of the algorithm in which the number of assumptions made in the present analysis are reduced. © 1996 by John Wiley & Sons, Inc.

No abstract