Aceleración de Algoritmos en Procesadores Multicore, GPGPU y FPGA. Algorithm Acceleration with Multicore, GPGPU and FPGA Processors.

El avance en la potencia de cómputo en nuestros días viene dado por la paralelización del procesamiento, dadas las características que disponen las nuevas arquitecturas de hardware. Utilizar convenientemente este hardware impacta en la aceleración de los algoritmos en ejecución (programas). Sin embargo, convertir de forma adecuada el algoritmo en su forma paralela es complejo, y a su vez, esta forma, es específica para cada tipo de hardware paralelo. En la actualidad los procesadores de uso general más comunes son los multicore, procesadores paralelos, también denominados Symmetric Multi-Processors (SMP). Hoy en día es difícil hallar un procesador para computadoras de escritorio que no tengan algún tipo de paralelismo del caracterizado por los SMP, siendo la tendencia de desarrollo, que cada día nos encontremos con procesadores con mayor numero de cores disponibles. Por otro lado, los dispositivos de procesamiento de video (Graphics Processor Units - GPU), a su vez, han ido desarrollando su potencia de cómputo por medio de disponer de múltiples unidades de procesamiento dentro de su composición electrónica, a tal punto que en la actualidad no es difícil encontrar placas de GPU con capacidad de 200 a 400 hilos de procesamiento paralelo. Estos procesadores son muy veloces y específicos para la tarea que fueron desarrollados, principalmente el procesamiento de video. Sin embargo, como este tipo de procesadores tiene muchos puntos en común con el procesamiento científico, estos dispositivos han ido reorientándose con el nombre de General Processing Graphics Processor Unit (GPGPU). A diferencia de los procesadores SMP señalados anteriormente, las GPGPU no son de propósito general y tienen sus complicaciones para uso general debido al límite en la cantidad de memoria que cada placa puede disponer y al tipo de procesamiento paralelo que debe realizar para poder ser productiva su utilización. Los dispositivos de lógica programable, FPGA, son dispositivos capaces de realizar grandes cantidades de operaciones en paralelo, por lo que pueden ser usados para la implementación de algoritmos específicos, aprovechando el paralelismo que estas ofrecen. Su inconveniente viene derivado de la complejidad para la programación y el testing del algoritmo instanciado en el dispositivo. Ante esta diversidad de procesadores paralelos, el objetivo de nuestro trabajo está enfocado en analizar las características especificas que cada uno de estos tienen, y su impacto en la estructura de los algoritmos para que su utilización pueda obtener rendimientos de procesamiento acordes al número de recursos utilizados y combinarlos de forma tal que su complementación sea benéfica. Específicamente, partiendo desde las características del hardware, determinar las propiedades que el algoritmo paralelo debe tener para poder ser acelerado. Las características de los algoritmos paralelos determinará a su vez cuál de estos nuevos tipos de hardware son los mas adecuados para su instanciación. En particular serán tenidos en cuenta el nivel de dependencia de datos, la necesidad de realizar sincronizaciones durante el procesamiento paralelo, el tamaño de datos a procesar y la complejidad de la programación paralela en cada tipo de hardware. Today´s advances in high-performance computing are driven by parallel processing capabilities of available hardware architectures. These architectures enable the acceleration of algorithms when thes ealgorithms are properly parallelized and exploit the specific processing power of the underneath architecture. Most current processors are targeted for general pruposes and integrate several processor cores on a single chip, resulting in what is known as a Symmetric Multiprocessing (SMP) unit. Nowadays even desktop computers make use of multicore processors. Meanwhile, the industry trend is to increase the number of integrated rocessor cores as technology matures. On the other hand, Graphics Processor Units (GPU), originally designed to handle only video processing, have emerged as interesting alternatives to implement algorithm acceleration. Current available GPUs are able to implement from 200 to 400 threads for parallel processing. Scientific computing can be implemented in these hardware thanks to the programability of new GPUs that have been denoted as General Processing Graphics Processor Units (GPGPU).However, GPGPU offer little memory with respect to that available for general-prupose processors; thus, the implementation of algorithms need to be addressed carefully. Finally, Field Programmable Gate Arrays (FPGA) are programmable devices which can implement hardware logic with low latency, high parallelism and deep pipelines. Thes devices can be used to implement specific algorithms that need to run at very high speeds. However, their programmability is harder that software approaches and debugging is typically time-consuming. In this context where several alternatives for speeding up algorithms are available, our work aims at determining the main features of thes architectures and developing the required know-how to accelerate algorithm execution on them. We look at identifying those algorithms that may fit better on a given architecture as well as compleme

An objective anisotropic elastic plastic model and Algorithm applicable to bone mechanics

The implicit projection algorithm of isotropic plasticity is extended to an objective anisotropic elastic perfectly plastic model. The recursion formula developed to project the trial stress on the yield surface, is applicable to any non linear elastic law and any plastic yield function.A curvilinear transverse isotropic model based on a quadratic elastic potential and on Hill's quadratic yield criterion is then developed and implemented in a computer program for bone mechanics perspectives. The paper concludes with a numerical study of a schematic bone-prosthesis system to illustrate the potential of the model.

Disease-causing mutations improving the branch site and polypyrimidine tract: pseudoexon activation of LINE-2 and antisense Alu lacking the poly(T)-tail.

Cryptic exons or pseudoexons are typically activated by point mutations that create GT or AG dinucleotides of new 5' or 3' splice sites in introns, often in repetitive elements. Here we describe two cases of tetrahydrobiopterin deficiency caused by mutations improving the branch point sequence and polypyrimidine tracts of repeat-containing pseudoexons in the PTS gene. In the first case, we demonstrate a novel pathway of antisense Alu exonization, resulting from an intronic deletion that removed the poly(T)-tail of antisense AluSq. The deletion brought a favorable branch point sequence within proximity of the pseudoexon 3' splice site and removed an upstream AG dinucleotide required for the 3' splice site repression on normal alleles. New Alu exons can thus arise in the absence of poly(T)-tails that facilitated inclusion of most transposed elements in mRNAs by serving as polypyrimidine tracts, highlighting extraordinary flexibility of Alu repeats in shaping intron-exon structure. In the other case, a PTS pseudoexon was activated by an A>T substitution 9 nt upstream of its 3' splice site in a LINE-2 sequence, providing the first example of a disease-causing exonization of the most ancient interspersed repeat. These observations expand the spectrum of mutational mechanisms that introduce repetitive sequences in mature transcripts and illustrate the importance of intronic mutations in alternative splicing and phenotypic variability of hereditary disorders.

Conversion of membrane-bound Fas(CD95) ligand to its soluble form is associated with downregulation of its proapoptotic activity and loss of liver toxicity.

Human Fas ligand (L) (CD95L) and tumor necrosis factor (TNF)-alpha undergo metalloproteinase-mediated proteolytic processing in their extracellular domains resulting in the release of soluble trimeric ligands (soluble [s]FasL, sTNF-alpha) which, in the case of sFasL, is thought to be implicated in diseases such as hepatitis and AIDS. Here we show that the processing of sFasL occurs between Ser126 and Leu127. The apoptotic-inducing capacity of naturally processed sFasL was reduced by >1,000-fold compared with membrane-bound FasL, and injection of high doses of recombinant sFasL in mice did not induce liver failure. However, soluble FasL retained its capacity to interact with Fas, and restoration of its cytotoxic activity was achieved both in vitro and in vivo with the addition of cross-linking antibodies. Similarly, the marginal apoptotic activity of recombinant soluble TNF-related apoptosis-inducing ligand (sTRAIL), another member of the TNF ligand family, was greatly increased upon cross-linking. These results indicate that the mere trimerization of the Fas and TRAIL receptors may not be sufficient to trigger death signals. Thus, the observation that sFasL is less cytotoxic than membrane-bound FasL may explain why in certain types of cancer, systemic tissue damage is not detected, even though the levels of circulating sFasL are high.

Structural prediction of peptides bound to MHC class I.

An ab initio structure prediction approach adapted to the peptide-major histocompatibility complex (MHC) class I system is presented. Based on structure comparisons of a large set of peptide-MHC class I complexes, a molecular dynamics protocol is proposed using simulated annealing (SA) cycles to sample the conformational space of the peptide in its fixed MHC environment. A set of 14 peptide-human leukocyte antigen (HLA) A0201 and 27 peptide-non-HLA A0201 complexes for which X-ray structures are available is used to test the accuracy of the prediction method. For each complex, 1000 peptide conformers are obtained from the SA sampling. A graph theory clustering algorithm based on heavy atom root-mean-square deviation (RMSD) values is applied to the sampled conformers. The clusters are ranked using cluster size, mean effective or conformational free energies, with solvation free energies computed using Generalized Born MV 2 (GB-MV2) and Poisson-Boltzmann (PB) continuum models. The final conformation is chosen as the center of the best-ranked cluster. With conformational free energies, the overall prediction success is 83% using a 1.00 Angstroms crystal RMSD criterion for main-chain atoms, and 76% using a 1.50 Angstroms RMSD criterion for heavy atoms. The prediction success is even higher for the set of 14 peptide-HLA A0201 complexes: 100% of the peptides have main-chain RMSD values < or =1.00 Angstroms and 93% of the peptides have heavy atom RMSD values < or =1.50 Angstroms. This structure prediction method can be applied to complexes of natural or modified antigenic peptides in their MHC environment with the aim to perform rational structure-based optimizations of tumor vaccines.

The E.coli RuvAB proteins branch migrate Holliday junctions through heterologous DNA sequences in a reaction facilitated by SSB.

During genetic recombination a heteroduplex joint is formed between two homologous DNA molecules. The heteroduplex joint plays an important role in recombination since it accommodates sequence heterogeneities (mismatches, insertions or deletions) that lead to genetic variation. Two Escherichia coli proteins, RuvA and RuvB, promote the formation of heteroduplex DNA by catalysing the branch migration of crossovers, or Holliday junctions, which link recombining chromosomes. We show that RuvA and RuvB can promote branch migration through 1800 bp of heterologous DNA, in a reaction facilitated by the presence of E.coli single-stranded DNA binding (SSB) protein. Reaction intermediates, containing unpaired heteroduplex regions bound by SSB, were directly visualized by electron microscopy. In the absence of SSB, or when SSB was replaced by a single-strand binding protein from bacteriophage T4 (gene 32 protein), only limited heterologous branch migration was observed. These results show that the RuvAB proteins, which are induced as part of the SOS response to DNA damage, allow genetic recombination and the recombinational repair of DNA to occur in the presence of extensive lengths of heterology.

Use of Hy's law and a new composite algorithm to predict acute liver failure in patients with drug-induced liver injury.

BACKGROUND & AIMS Hy's Law, which states that hepatocellular drug-induced liver injury (DILI) with jaundice indicates a serious reaction, is used widely to determine risk for acute liver failure (ALF). We aimed to optimize the definition of Hy's Law and to develop a model for predicting ALF in patients with DILI. METHODS We collected data from 771 patients with DILI (805 episodes) from the Spanish DILI registry, from April 1994 through August 2012. We analyzed data collected at DILI recognition and at the time of peak levels of alanine aminotransferase (ALT) and total bilirubin (TBL). RESULTS Of the 771 patients with DILI, 32 developed ALF. Hepatocellular injury, female sex, high levels of TBL, and a high ratio of aspartate aminotransferase (AST):ALT were independent risk factors for ALF. We compared 3 ways to use Hy's Law to predict which patients would develop ALF; all included TBL greater than 2-fold the upper limit of normal (×ULN) and either ALT level greater than 3 × ULN, a ratio (R) value (ALT × ULN/alkaline phosphatase × ULN) of 5 or greater, or a new ratio (nR) value (ALT or AST, whichever produced the highest ×ULN/ alkaline phosphatase × ULN value) of 5 or greater. At recognition of DILI, the R- and nR-based models identified patients who developed ALF with 67% and 63% specificity, respectively, whereas use of only ALT level identified them with 44% specificity. However, the level of ALT and the nR model each identified patients who developed ALF with 90% sensitivity, whereas the R criteria identified them with 83% sensitivity. An equal number of patients who did and did not develop ALF had alkaline phosphatase levels greater than 2 × ULN. An algorithm based on AST level greater than 17.3 × ULN, TBL greater than 6.6 × ULN, and AST:ALT greater than 1.5 identified patients who developed ALF with 82% specificity and 80% sensitivity. CONCLUSIONS When applied at DILI recognition, the nR criteria for Hy's Law provides the best balance of sensitivity and specificity whereas our new composite algorithm provides additional specificity in predicting the ultimate development of ALF.

New electrocardiographic criteria for discriminating between Brugada types 2 and 3 patterns and incomplete right bundle branch block.

OBJECTIVES: The aim of this study was to evaluate new electrocardiographic (ECG) criteria for discriminating between incomplete right bundle branch block (RBBB) and the Brugada types 2 and 3 ECG patterns. BACKGROUND: Brugada syndrome can manifest as either type 2 or type 3 pattern. The latter should be distinguished from incomplete RBBB, present in 3% of the population. METHODS: Thirty-eight patients with either type 2 or type 3 Brugada pattern that were referred for an antiarrhythmic drug challenge (AAD) were included. Before AAD, 2 angles were measured from ECG leads V(1) and/or V(2) showing incomplete RBBB: 1) α, the angle between a vertical line and the downslope of the r'-wave, and 2) β, the angle between the upslope of the S-wave and the downslope of the r'-wave. Baseline angle values, alone or combined with QRS duration, were compared between patients with negative and positive results on AAD. Receiver-operating characteristic curves were constructed to identify optimal discriminative cutoff values. RESULTS: The mean β angle was significantly smaller in the 14 patients with negative results on AAD compared to the 24 patients with positive results on AAD (36 ± 20° vs. 62 ± 20°, p < 0.01). Its optimal cutoff value was 58°, which yielded a positive predictive value of 73% and a negative predictive value of 87% for conversion to type 1 pattern on AAD; α was slightly less sensitive and specific compared with β. When the angles were combined with QRS duration, it tended to improve discrimination. CONCLUSIONS: In patients with suspected Brugada syndrome, simple ECG criteria can enable discrimination between incomplete RBBB and types 2 and 3 Brugada patterns.

Algorithm creation and optimization for the crew pairing problem

Algoritmo que optimiza y crea pairings para tripulaciones de líneas aéreas mediante la posterior programación en Java.