No abstract

A newly developed high-throughput desorption electrospray ionization (DESI) source was characterized in terms of its performance in quantitative analysis. A 96-sample array, containing pharmaceuticals in various matrices, was analyzed in a single run with a total analysis time of 3 min. These solution-phase samples were examined from a hydrophobic PTFE ink printed on glass. The quantitative accuracy, precision, and limit of detection (LOD) were characterized. Chemical background-free samples of propranolol (PRN) with PRN-d 7 as internal standard (IS) and carbamazepine (CBZ) with CBZ-d 10 as IS were examined. So were two other sample sets consisting of PRN/PRN-d 7 at varying concentration in a biological milieu of 10% urine or porcine brain total lipid extract, total lipid concentration 250 ng/ L. The background-free samples, examined in a total analysis time of 1.5 s/sample, showed good quantitative accuracy and precision, with a relative error (RE) and relative standard deviation (RSD) generally less than 3% and 5%, respectively. The samples in urine and the lipid extract required a longer analysis time (2.5 s/sample) and showed RSD values of around 10% for the samples in urine and 4% for the lipid extract samples and RE values of less than 3% for both sets. The LOD for PRN and CBZ when analyzed without chemical background was 10 and 30 fmol, respectively. The LOD of PRN increased to 400 fmol analyzed in 10% urine, and 200 fmol when analyzed in the brain lipid extract. (J Am Soc Mass Spectrom 2009, 20, 321-325) © 2009 Published by Elsevier Inc. on behalf of American Society for Mass Spectrometry Q uantitative analysis of pharmaceutical compounds in biological fluids is typically carried out using liquid chromatography separation combined with a mass spectrometer fitted with an atmospheric pressure ionization (API) ion source, usually electrospray ionization or atmospheric pressure chemical ionization (LC/ESI-MS or LC/APCI-MS). Such a system can analyze one sample for numerous components in a matter of a few minutes [1]. In most applications, the separation device is directly coupled on-line with the MS. In others, however, extensive prior separation is not necessary, and it can be replaced by a simpler extraction step. In these cases, current highthroughput approaches include direct injection of the sample solution into the MS from a 96-well sample plate for analysis by API [2], or the spotting of solid samples with an organic matrix for analysis by matrix-assisted laser desorption/ionization (MALDI) [3].New options for the ionization of samples at atmospheric pressure have emerged recently. The family of ambient ionization methods for mass spectrometry, which originated with desorption electrospray ionization (DESI) [4] and direct analysis in real time (DART) [5], has rapidly expanded to include a number of new members including MALDESI, LAESI, PADI, ASAP, DBDI, and ELDI [6 -11]. In DESI-MS, samples are analyzed by directing a pneumatically-assisted stream of charged microdroplets at the surface in the ambient...

Much of the software in everyday operation is not making optimal use of the hardware on which it actually runs. Among the reasons for this discrepancy are hardware/software mismatches, modularization overheads introduced by software engineering considerations, and the inability of systems to adapt to users' behaviors.A solution to these problems is to delay code generation until load time. This is the earliest point at which a piece of software can be fine-tuned to the actual capabilities of the hardware on which it is about to be executed, and also the earliest point at wich modularization overheads can be overcome by global optimization.A still better match between software and hardware can be achieved by replacing the already executing software at regular intervals by new versions constructed on-the-fly using a background code re-optimizer. This not only enables the use of live profiling data to guide optimization decisions, but also facilitates adaptation to changing usage patterns and the late addition of dynamic link libraries.This paper presents a system that provides code generation at load-time and continuous program optimization at run-time. First, the architecture of the system is presented. Then, two optimization techniques are discussed that were developed specifically in the context of continuous optimization. The first of these optimizations continually adjusts the storage layouts of dynamic data structures to maximize data cache locality, while the second performs profiledriven instruction re-scheduling to increase instruction-level parallelism. These two optimizations have very different cost/benefit ratios, presented in a series of benchmarks. The paper concludes with an outlook to future research directions and an enumeration of some remaining research problems.The empirical results presented in this paper make a case in favor of continuous optimization, but indicate that it needs to be applied judiciously. In many situations, the costs of dynamic optimizations outweigh their benefit, so that no break-even point is ever reached. In favorable circumstances, on the other hand, speed-ups of over 120% have been observed. It appears as if the

In this paper we introduce a programming language for Web document processing called WebL. WebL is a high level, object-oriented scripting language that incorporates two novel features: service combinators and a markup algebra. Service combinators are language constructs that provide reliable access to web services by mimicking a web surfer's behavior when a failure occurs while retrieving a page. The markup algebra extracts structured and unstructured values from pages for computation, and is based on algebraic operations on sets of markup elements. WebL is used to quickly build and experiment with custom web crawlers, meta-search engines, page transducers, shopping robots, etc.

We present and evaluate a simple, yet efficient optimization technique that improves memoryhierarchy performance for pointer-centric applications by up to 24% and reduces cache misses by up to 35%. This is achieved by selecting an improved ordering for the data members of pointerbased data structures. Our optimization is applicable to all type-safe programming languages that completely abstract from physical storage layout; examples of such languages are Java and Oberon. Our technique does not involve programmers in the optimization process, but runs fully automatically, guided by dynamic profiling information that captures which paths through the program are taken with what frequency. The algorithm first strives to cluster data members that are accessed closely after one another onto the same cache line, increasing spatial locality. Then, the data members that have been mapped to a particular cache line are ordered to minimize load latency in case of a cache miss.