Adam Hughes

Exploiting multiple levels of parallelism in Molecular Dynamics based calculations via modern techniques and software paradigms on distributed memory computers

Abstract Modern Molecular Dynamics methods are employed to study quantum manybody systems, chemically reactive systems including explicit electronic degrees of freedom, and combinations thereof, as well as large classical biomolecular systems. Thus, complex problems such as isotope effects on enzymatic reactions can now be examined, routinely. In this article, modern molecular dynamics methods are reviewed and their application to quantum manybody systems and electronic structure calculations described. The resulting methodology, however, while powerful, is computationally intensive. Therefore, the mathematical structure of the techniques has been exploited to develop distributed memory parallel algorithms employing multiple levels of discretization. These multilevel-parallel methods are efficient and permit the large complex systems, such as enzyme catalysis, to be treated easily. In addition, it is shown how modern object oriented programming paradigms can be employed to implement multilevel parallel algorithms in a large computational package rapidly and efficiently. Finally, results and timings obtaining using the PINY_MD package developed by the authors are given for a variety of novel systems.

Molecular dynamics algorithms for path integrals at constant pressure

Extended system path integral molecular dynamics algorithms have been developed that can generate efficiently averages in the quantum mechanical canonical ensemble [M. E. Tuckerman, B. J. Berne, G. J. Martyna, and M. L. Klein, J. Chem. Phys. 99, 2796 (1993)]. Here, the corresponding extended system path integral molecular dynamics algorithms appropriate to the quantum mechanical isothermal–isobaric ensembles with isotropic-only and full system cell fluctuations are presented. The former ensemble is employed to study fluid systems which do not support shear modes while the latter is employed to study solid systems. The algorithms are constructed by deriving appropriate dynamical equations of motions and developing reversible multiple time step algorithms to integrate the equations numerically. Effective parallelization schemes for distributed memory computers are presented. The new numerical methods are tested on model (a particle in a periodic potential) and realistic (liquid and solid para-hydrogen and liquid butane) systems. In addition, the methodology is extended to treat the path integral centroid dynamics scheme, [J. Cao and G. A. Voth, J. Chem. Phys. 99, 10070 (1993)], a novel method which is capable of generating semiclassical approximations to quantum mechanical time correlation functions.

Hybrid cloud and cluster computing paradigms for life science applications

BackgroundClouds and MapReduce have shown themselves to be a broadly useful approach to scientific computing especially for parallel data intensive applications. However they have limited applicability to some areas such as data mining because MapReduce has poor performance on problems with an iterative structure present in the linear algebra that underlies much data analysis. Such problems can be run efficiently on clusters using MPI leading to a hybrid cloud and cluster environment. This motivates the design and implementation of an open source Iterative MapReduce system Twister.ResultsComparisons of Amazon, Azure, and traditional Linux and Windows environments on common applications have shown encouraging performance and usability comparisons in several important non iterative cases. These are linked to MPI applications for final stages of the data analysis. Further we have released the open source Twister Iterative MapReduce and benchmarked it against basic MapReduce (Hadoop) and MPI in information retrieval and life sciences applications.ConclusionsThe hybrid cloud (MapReduce) and cluster (MPI) approach offers an attractive production environment while Twister promises a uniform programming environment for many Life Sciences applications.MethodsWe used commercial clouds Amazon and Azure and the NSF resource FutureGrid to perform detailed comparisons and evaluations of different approaches to data intensive computing. Several applications were developed in MPI, MapReduce and Twister in these different environments.

Applying Twister to Scientific Applications

Many scientific applications suffer from the lack of a unified approach to support the management and efficient processing of large-scale data. The Twister MapReduce Framework, which not only supports the traditional MapReduce programming model but also extends it by allowing iterations, addresses these problems. This paper describes how Twister is applied to several kinds of scientific applications such as BLAST, MDS Interpolation and GTM Interpolation in a non-iterative style and to MDS without interpolation in an iterative style. The results show the applicability of Twister to data parallel and EM algorithms with small overhead and increased efficiency.

Constant pressure path integral molecular dynamics studies of quantum effects in the liquid state properties of n-alkanes

A computer simulation study of quantum effects in methane, butane, and octane is presented. Each molecular system is examined at three state points in the liquid region using novel extended system, multiple time step, constant pressure, path integral molecular dynamics methodology. In addition, the results of classical calculations are reported to provide a useful reference. Liquid butane is used as a test case on which to compare the predictions of two empirical force fields, CHARMM22 and AMBER95. Comparisons are made to experiment. Briefly, the models predict that quantum effects lead to an increase in molar volume of approximately 2 cm3/mole (i.e., relative to a classical calculation). However, a slight unphysical hydrogen–deuterium isotope effect is, also, observed. This may be caused by an incorrect parametrization of the anisotropy of the potential or by a reduction in the magnitude of the intermolecular induced dipole-induced dipole dispersion coefficient with increasing isotope mass that has not been parametrized in the force fields. In addition, the results show an interesting zero-point energy effect. The intramolecular regions of the radial distribution function exhibit less structure at lower temperatures than at higher temperatures. This is the inverse of the prediction of the model in the classical limit. The quantum effect occurs because the bulk density decreases faster than the intramolecular degrees of freedom lose zero-point energy as temperature increases in the highly harmonic intramolecular potential model employed in the calculations. Nonetheless, the phenomena is not likely to be an artifact and careful experiments could observe it. Finally, the efficiency of the path molecular dynamics methods employed in the studies are demonstrated on both serial and parallel computers.

Interpolative multidimensional scaling techniques for the identification of clusters in very large sequence sets

BackgroundModern pyrosequencing techniques make it possible to study complex bacterial populations, such as 16S rRNA, directly from environmental or clinical samples without the need for laboratory purification. Alignment of sequences across the resultant large data sets (100,000+ sequences) is of particular interest for the purpose of identifying potential gene clusters and families, but such analysis represents a daunting computational task. The aim of this work is the development of an efficient pipeline for the clustering of large sequence read sets.MethodsPairwise alignment techniques are used here to calculate genetic distances between sequence pairs. These methods are pleasingly parallel and have been shown to more accurately reflect accurate genetic distances in highly variable regions of rRNA genes than do traditional multiple sequence alignment (MSA) approaches. By utilizing Needleman-Wunsch (NW) pairwise alignment in conjunction with novel implementations of interpolative multidimensional scaling (MDS), we have developed an effective method for visualizing massive biosequence data sets and quickly identifying potential gene clusters.ResultsThis study demonstrates the use of interpolative MDS to obtain clustering results that are qualitatively similar to those obtained through full MDS, but with substantial cost savings. In particular, the wall clock time required to cluster a set of 100,000 sequences has been reduced from seven hours to less than one hour through the use of interpolative MDS.ConclusionsAlthough work remains to be done in selecting the optimal training set size for interpolative MDS, substantial computational cost savings will allow us to cluster much larger sequence sets in the future.

Visualizing the Protein Sequence Universe

Modern biology is experiencing a rapid increase in data volumes that challenges our analytical skills and existing cyberinfrastructure. Exponential expansion of the protein sequence universe (PSU), the protein sequence space, together with the costs and complexities of manual curation creates a major bottleneck in life sciences research. Existing resources lack scalable visualization tools that are instrumental for functional annotation. Here, we describe a new visualization tool using multidimensional scaling to create a 3D embedding of the protein space. The advantages of the proposed PSU method include the ability to scale to large numbers of sequences, integrate different similarity measures with other functional and experimental data, and facilitate protein annotation. We applied the method to visualize the prokaryotic PSU using sequence alignment scores. As an annotation example, we used the interpolation approach to map the set of annotated archaeal proteins into the prokaryotic PSU. Transdisciplinary approaches akin to the one described in this paper are urgently needed to quickly and efficiently translate the influx of new data into tangible innovations and groundbreaking discoveries. Copyright

Visualizing the protein sequence universe

Modern biology is experiencing a rapid increase in data volumes that challenges our analytical skills and existing cyberinfrastructure. Exponential expansion of the Protein Sequence Universe (PSU), the protein sequence space, together with the costs and complexities of manual curation creates a major bottleneck in life sciences research. Existing resources lack scalable visualization tools that are instrumental for functional annotation. Here, we describe a multi-dimensional scaling (MDS) implementation to create a 3D embedding of the PSU that allows visualizing the relationships between large numbers of proteins. To demonstrate the method, we use sequence similarity scores as a measure of proximity. An example of the prokaryotic PSU shows that the low-dimensional representation preserves important grouping features such as relative proximity of functionally similar clusters and clear structural separation between clusters with specific and general functions. The advantages of the method and its implementation include the ability to scale to large numbers of sequences, integrate different similarity measures with other functional and experimental data, and facilitate protein annotation. Transdisciplinary approaches akin to the one described in this paper are urgently needed to quickly and efficiently translate the influx of new data into tangible innovations and groundbreaking discoveries.