TechTalks from event: IEEE IPDPS 2011

Note 1: Only plenary sessions (keynotes, panels, and best papers) are accessible without requiring log-in. For other talks, you will need to log-in using the email you registered for IPDPS 2011. Note 2: Many of the talks (those without a thumbnail next to the their description below) are yet to be uploaded. Some of them were not recorded because of technical problems. We are working with the corresponding authors to upload the self-recorded versions here. We sincerely thank all authors for their efforts in making their videos available.

SESSION 27: Computational Biology and Simulations

  • Smith-Waterman Alignment of Huge Sequences with GPU in Linear Space Authors: Edans Flavius de Oliveira Sandes (University of Brasilia, Brazil); Alba Cristina Magalhaes Alves de Melo (University of Brasili
    Cross-species chromosome alignments can reveal ancestral relationships and may be used to identify the peculiarities of the species. It is thus an important problem in Bioinformatics. So far, aligning huge sequences, such as whole chromosomes, with exact methods has been regarded as unfeasible, due to huge computing and memory requirements. However, high performance computing platforms such as GPUs are being able to change this scenario, making it possible to obtain the exact result for huge sequences in reasonable time. In this paper, we propose and evaluate a parallel algorithm that uses GPU to align huge sequences, executing the Smith-Waterman algorithm combined with Myers-Miller, with linear space complexity. In order to achieve that, we propose optimizations that are able to reduce signi?cantly the amount of data processed and that enforce full parallelism most of the time. Using the GTX 285 Board, our algorithm was able to produce the optimal alignment between sequences composed of 33 Millions of Base Pairs (MBP) and 47 MBP in 18.5 hours.
  • Accelerating Protein Sequence Search in a Heterogeneous Computing System Authors: Shucai Xiao (Virginia Tech, USA); Heshan Lin (Virginia Tech, USA); Wu-chun Feng (Virginia Tech, USA)
    The “Basic Local Alignment Search Tool” (BLAST) is arguably the most widely used computational tool in bioinformatics. However, the computational power required for routine BLAST analysis has been outstripping Moore’s Law due to the exponential growth in the size of the genomic sequence databases that BLAST searches on. To address the above issue, we propose the design and optimization of the BLAST algorithm for searching protein sequences (i.e., BLASTP) in a heterogeneous computing system. The end result is a BLASTP implementation that delivers a seven-fold speedup over the sequential BLASTP for the most computationally intensive phase (i.e., hit detection and ungapped extension) on a NVIDIA Fermi C2050 GPU. In addition, when pipelining the processing on a dual-core CPU and the NVIDIA Fermi GPU, our implementation can achieve a six-fold speedup for the overall program execution.
  • Parallel Metagenomic Sequence Clustering via Sketching and Quasi-clique Enumeration on Map-reduce Clouds Authors: Xiao Yang (Iowa State University, USA); Jaroslaw Zola (Iowa State University, USA); Srinivas Aluru (Iowa State University, USA)
    Taxonomic clustering of species is an important and frequently arising problem in metagenomics. High-throughput next generation sequencing is facilitating the creation of large metagenomic samples, while at the same time making the clustering problem harder due to the short sequence length supported and unknown species sampled. In this paper, we present a parallel algorithm for hierarchical taxonomic clustering of large metagenomic samples with support for overlapping clusters. We adapt the sketching techniques originally developed for web document clustering to deduce signi?cant similarities between pairs of sequences without resorting to expensive all vs. all alignments. We formulate the metagenomics classi?cation problem as that of maximal quasi-clique enumeration in the resulting similarity graph, at multiple levels of the hierarchy as prescribed by different similarity thresholds. We cast execution of the underlying algorithmic steps as applications of the map-reduce framework to achieve a cloud based implementation. Apart from solving an important problem in metagenomics, this work demonstrates the applicability of map-reduce framework in relatively complicated algorithmic settings.
  • Large-scale lattice gas Monte Carlo simulations for the generalized Ising model Authors: Tobias C. Kerscher (Technische Universität Hamburg-Harburg, Germany); Stefan Müller (Technische Universität Hambu
    We present an ef?cient parallel algorithm for lattice gas Monte Carlo simulations in the framework of an Ising model that allows arbitrary interaction on any lattice, a model often called a cluster expansion. Thermodynamic Monte Carlo simulations strive for the equilibrium properties of a system by exchanging atoms over a long range, while preserving detailed balance. This long-range exchange of atoms renders other frequent parallelization techniques, like domain decomposition, unfavorable due to excessive communication cost. Our ansatz, based on the Metropolis algorithm, minimizes communication between parallel processes. We present this new “partial sequence preserving” (PSP) algorithm, as well as benchmark data for a physical alloy system (NiAl) comprised of one billion atoms.