Efficiently exploring architectural design spaces via predictive modeling

Architects use cycle-by-cycle simulation to evaluate design choices and understand tradeoffs and interactions among design parameters. Efficiently exploring exponential-size design spaces with many interacting parameters remains an open problem: the sheer number of experiments renders detailed simulation intractable. We attack this problem via an automated approach that builds accurate, confident predictive design-space models. We simulate sampled points, using the results to teach our models the function describing relationships among design parameters. The models produce highly accurate performance estimates for other points in the space, can be queried to predict performance impacts of architectural changes, and are very fast compared to simulation, enabling efficient discovery of tradeoffs among parameters in different regions. We validate our approach via sensitivity studies on memory hierarchy and CPU design spaces: our models generally predict IPC with only 1-2% error and reduce required simulation by two orders of magnitude. We also show the efficacy of our technique for exploring chip multiprocessor (CMP) design spaces: when trained on a 1% sample drawn from a CMP design space with 250K points and up to 55x performance swings among different system configurations, our models predict performance with only 4-5% error on average. Our approach combines with techniques to reduce time per simulation, achieving net time savings of three-four orders of magnitude. Copyright © 2006 ACM.

NuGO contributions to GenePattern

NuGO, the European Nutrigenomics Organization, utilizes 31 powerful computers for, e.g., data storage and analysis. These so-called black boxes (NBXses) are located at the sites of different partners. NuGO decided to use GenePattern as the preferred genomic analysis tool on each NBX. To handle the custom made Affymetrix NuGO arrays, new NuGO modules are added to GenePattern. These NuGO modules execute the latest Bioconductor version ensuring up-to-date annotations and access to the latest scientific developments. The following GenePattern modules are provided by NuGO: NuGOArrayQualityAnalysis for comprehensive quality control, NuGOExpressionFileCreator for import and normalization of data, LimmaAnalysis for identification of differentially expressed genes, TopGoAnalysis for calculation of GO enrichment, and GetResultForGo for retrieval of information on genes associated with specific GO terms. All together, these NuGO modules allow comprehensive, up-to-date, and user friendly analysis of Affymetrix data. A special feature of the NuGO modules is that for analysis they allow the use of either the standard Affymetrix or the MBNI custom CDF-files, which remap probes based on current knowledge. In both cases a .chip-file is created to enable GSEA analysis. The NuGO GenePattern installations are distributed as binary Ubuntu (.deb) packages via the NuGO repository.

Strategies to Improve the Surface Plasmon Resonance-Based Immmunodetection of Bacterial Cells

We have made a comparison of (a) different surface chemistries of surface plasmon resonance (SPR) sensor chips (such as carboxymethylated dextran and carboxymethylated C1) and (b) of different assay formats (direct, sandwich and subtractive immunoassay) in order to improve the sensitivity of the determination of the model bacteria Acidovorax avenae subsp. citrulli (Aac). The use of the carboxymethylated sensor chip C1 resulted in a better sensitivity than that of carboxymethylated dextran CM5 in all the assay formats. The direct assay format, in turn, exhibits the best sensitivity. Thus, the combination of a carboxymethylated sensor chip C1 with the direct assay format resulted in the highest sensitivity for Aac, with a limit of detection of 1.6x106 CFU mL-1. This SPR immunosensor was applied to the detection of Aac in watermelon leaf extracts spiked with the bacteria, and the lower LOD is 2.2x107 CFU mL-1.