What is the current state of the art in protein design? This question was approached in a recent two-week protein design workshop sponsored by EMBO and held at the EMBL in Heidelberg. The goals were to test available ...
详细信息
What is the current state of the art in protein design? This question was approached in a recent two-week protein design workshop sponsored by EMBO and held at the EMBL in Heidelberg. The goals were to test available design tools and to explore new design strategies. Five novel proteins were designed: Shpilka, a sandwich of two four-stranded beta-sheets, a scaffold on which to explore variations in loop topology;Grendel, a four-helical membrane anchor, ready for fusion to water-soluble functional domains;Finger-clasp, a dimer of interdigitating beta-beta-alpha-units, the simplest variant of the "handshake" structural class;Aida, an antibody binding surface intended to be specific for flavodoxin;Leather-a minimal NAD binding domain, extracted from a larger protein. Each design is available as a set of three-dimensional coordinates, the corresponding amino acid sequence and a set of analytical results. The designs are placed in the public domain for scrutiny, improvement, and possible experimental verification.
A new method for large-scale numerical simulations of neural networks is proposed which reduces the computational effort by incrementally updating the local fields and thus restricting the operations to flipped spins ...
详细信息
A new method for large-scale numerical simulations of neural networks is proposed which reduces the computational effort by incrementally updating the local fields and thus restricting the operations to flipped spins only. A highly optimized multi-spin algorithm is described employing words oriented along the columns of the coupling matrix unlike the horizontal structure in existing high-speed algorithms. An effective rate of 35+109 couplings/s on a Cray-YMP can be attained which is about five times as fast as best existing multi-spin implementations.
Methods are given for computing extended maximum likelihood estimates in which one or more parameter estimates are infinite at the supremum of the likelihood. The results are given for a broad class of regression-like...
详细信息
Methods are given for computing extended maximum likelihood estimates in which one or more parameter estimates are infinite at the supremum of the likelihood. The results are given for a broad class of regression-like models based on independent observations with linearly related parameters including, in particular, the generalized linear models. The estimation consists of two steps: A linear programming step to identify the infinite components and a more conventional function optimization step to optimize the remaining finite components. Provision is made for nuisance parameters. Two algorithms are presented and examples illustrating their use are given.
As a result of an analysis of data relating tissue response to radiation absorbed dose the ICRU has recommended a target for accuracy of .+-. 5 for dose delivery in radiation therapy. This is a difficult overall objec...
详细信息
As a result of an analysis of data relating tissue response to radiation absorbed dose the ICRU has recommended a target for accuracy of .+-. 5 for dose delivery in radiation therapy. This is a difficult overall objective to achieve because of the many steps that make up a course of radiotherapy. The calculation of absorbed dose is only one of the steps and so to achieve an overall accuracy of better than .+-. 5% the accuracy in dose calculation must be better yet. The physics behind the problem is sufficiently complicated so that no exact method of calculation has been found and consequently approximate solutions must be used. The development of computer algorithms for this task involves the search for better and better approximate solutions. To achieve the desired target of accuracy a fairly sophisticated calculation procedure must be used. Only when this is done can we hope to further improve our knowledge of the way in which tissues respond to radiation treatments.
This paper applies mathematical ideas which arise from differential geometry and control theory to computer programming theory. For the sake of simplicity, most questions of logic and language are put to the side. A c...
详细信息
This paper applies mathematical ideas which arise from differential geometry and control theory to computer programming theory. For the sake of simplicity, most questions of logic and language are put to the side. A computer program is regarded as a collection of data, considered as a collection of cross-sections of a fiber space whose base is a discrete space. These distinguished cross-sections are called the trajectories of the computer program system. For the simplest von Neumann type of program, the base space is the nonnegative integers: The trajectories may then often be thought of as the orbits of a discretetime dynamical system. Feedback control theory enters when one attempts to provide a ‘rational’ way of generating such dynamical systems, satisfying certain conditions, such as termination and correctness. The greatest common divisor and bubble sort programs — standard examples in computer science textbooks — are treated as models for the development of a general theory. Geometrically, we will also consider them as the discrete analogue of what are called path systems in differential geometry. These examples can also be treated as the analogue of closed-loop feedback control systems — themselves a sort of path system — that are ‘driven’ to the final answer by a process that is similar to the way that Lyapunov theory is used in the linear regulator problem of control theory. This use of the Lyapunov ideas of dynamical system theory gives a technique for proving the correctness and termination of some programs. In order to provide a mathematical structure to a programming theory that is, in some sense, analogous to the group-theoretic structure of fiber bundle differential geometry, an algebraic structure of a category which acts on the fiber space is associated with such programs. The computation of the binomial coefficients will also be treated as an example of a system of partial difference equations, leading to the description of some features of concurr
A crucial concern when implementing computer algorithms for geostatistical analyses on microcomputers is speed of execution. Kriging, in particular, typically relies on a Gauss elimination algorithm to solve for weigh...
详细信息
The current state of large scale, numerical simulations of neural networks is reviewed. Hardware and software improvements make it likely that biological size networks, i.e., networks with more than 10 10 couplings, c...
详细信息
The current state of large scale, numerical simulations of neural networks is reviewed. Hardware and software improvements make it likely that biological size networks, i.e., networks with more than 10 10 couplings, can be simulated in the near future. Sample programs for the efficient simulation of a few simple models are presented as an aid to researchers just entering the field.
This paper presents a multimodal network design model with a computer algorithm to solve it. The selection of link improvements in the network is assumed to affect practical capacities as well as mean free times. The ...
详细信息
This paper presents a multimodal network design model with a computer algorithm to solve it. The selection of link improvements in the network is assumed to affect practical capacities as well as mean free times. The equilibrium flows and the modal shares are affected accordingly. The improvements are assumed to be continuous variables with concave marginal costs. This provided a good approximation for the behaviour of the discrete model. The algorithm is based on Hooke and Jeeves' Optimization Methodology which requires a solution of a multimodal equilibrium assignment problem for each functional evaluation. The algorithm is implemented in such a way that only some functional evaluations need a solution of an equilibrium problem, implying efficiency for large-scale problems. Computational results show that although flow congestion decreased due to link improvements, modal shares may change towards private users. The latter contributes to congestion again and possibly makes the overall decrease in congestion less significant.
The development of two computer algorithms to assess the performance of IDMT over-current protection for interconnected power systems is described. The first algorithm considers the case of three-phase short-circuit f...
详细信息
The development of two computer algorithms to assess the performance of IDMT over-current protection for interconnected power systems is described. The first algorithm considers the case of three-phase short-circuit faults, and the second considers single-phase to earth shortcircuit faults. Both can include multiple bulk supply points, parallel lines and transformers, directional and nondirectional relays, including both on the same circuit, teed points, and instantaneous high-set elements. The algorithms currently incorporate relays with a standard inverse characteristic but any relay with a current/time characteristic can easily be included. Results are presented of the application of these algorithms to the assessment of the relay performance for several actual power distribution systems. For each selected fault position, the operating time of each relay is calculated and the sequence of relay operation shown until the fault is completely isolated. The results also illustrate, in particular, the ways in which the performance of the relay systems can be tabulated to greatest effect.
A need exists for a set of computer programs that can be used by students to solve elementary digital signal processing problems using a personal computer. This project involved the design and implementation of ten al...
详细信息
A need exists for a set of computer programs that can be used by students to solve elementary digital signal processing problems using a personal computer. This project involved the design and implementation of ten algorithms that solve such problems and an additional algorithm that creates plots of the various input and output sequences. The two primary goals of the programs were: 1) user friendliness and, 2) portability. With these goals in mind, the source code was written using Fortran-77 and compiled by a commercially available Fortran compiler specifically designed for personal computers. The plotting program uses a Fortran-compatible graphics package that is also com- mercially available. The programs, once compiled, can be distributed to users without the requirement to purchase either a Fortran compiler or a graphics package; however, access to a Fortran compiler enhances the utility of the programs.
暂无评论