GPU-MEME: Using Graphics Hardware to Accelerate Motif Finding in DNA Sequences

Chen, Chen; Schmidt, Bertil; Liu, Weiguo; Müller‐Wittig, Wolfgang

doi:10.1007/978-3-540-88436-1_38

Cited by 16 publications

(6 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Multiple Expectation Maximization for Motif Elicitation (MEME) is a popular and efficient approximate algorithm used by researchers. Chen et al [10] have accelerated this algorithm using GPUs and have achieved significant speed up over its parallel version. The authors claim that more speed up can be achieved by using a cluster of GPUs.…”

Section: Related Workmentioning

confidence: 99%

Accelerating motif finding in DNA sequences with multicore CPUs

Perera

Ragel

2013

2013 IEEE 8th International Conference on Industrial and Information Systems

View full text Add to dashboard Cite

Motif discovery in DNA sequences is a challenging task in molecular biology. In computational motif discovery, Planted (l, d) motif finding is a widely studied problem and numerous algorithms are available to solve it. Both hardware and software accelerators have been introduced to accelerate the motif finding algorithms. However, the use of hardware accelerators such as FPGAs needs hardware specialists to design such systems. Software based acceleration methods on the other hand are easier to implement than hardware acceleration techniques. Grid computing is one such software based acceleration technique which has been used in acceleration of motif finding. However, drawbacks such as network communication delays and the need of fast interconnection between nodes in the grid can limit its usage and scalability. As using multicore CPUs to accelerate CPU intensive tasks are becoming increasingly popular and common nowadays, we can employ it to accelerate motif finding and it can be a faster method than grid based acceleration. In this paper, we have explored the use of multicore CPUs to accelerate motif finding. We have accelerated the Skip-Brute Force algorithm on multicore CPUs parallelizing it using the POSIX thread library. Our method yielded an average speed up of 34x on a 32-core processor compared to a speed up of 21x on a grid based implementation of 32 nodes

show abstract

Section: Related Workmentioning

confidence: 99%

Accelerating motif finding in DNA sequences with multicore CPUs

Perera

Ragel

2013

2013 IEEE 8th International Conference on Industrial and Information Systems

View full text Add to dashboard Cite

show abstract

“…Because these GPUs do not have to perform many of the generalized tasks that a CPU must perform, they have become highly optimized to perform tightly-coupled data-parallel processing with many, typically hundreds, of independent processor units and specialized memory addressing. GPU algorithms have been developed for many years for computational geometry tasks as part of graphics rendering, but it is only in the past few years where GPUs have been used for other tasks such as sequence analysis [ 12 - 14 ], machine learning [ 15 ] and molecular dynamics [ 16 , 17 ]. All of the early implementations had to contend with the constraints and difficulties of the limited programming environment available on the GPU, however this has changed in just the past couple of years.…”

Section: Introductionmentioning

confidence: 99%

Integrative multicellular biological modeling: a case study of 3D epidermal development using GPU algorithms

et al. 2010

View full text Add to dashboard Cite

Background: Simulation of sophisticated biological models requires considerable computational power. These models typically integrate together numerous biological phenomena such as spatially-explicit heterogeneous cells, cell-cell interactions, cell-environment interactions and intracellular gene networks. The recent advent of programming for graphical processing units (GPU) opens up the possibility of developing more integrative, detailed and predictive biological models while at the same time decreasing the computational cost to simulate those models. Results: We construct a 3D model of epidermal development and provide a set of GPU algorithms that executes significantly faster than sequential central processing unit (CPU) code. We provide a parallel implementation of the subcellular element method for individual cells residing in a lattice-free spatial environment. Each cell in our epidermal model includes an internal gene network, which integrates cellular interaction of Notch signaling together with environmental interaction of basement membrane adhesion, to specify cellular state and behaviors such as growth and division. We take a pedagogical approach to describing how modeling methods are efficiently implemented on the GPU including memory layout of data structures and functional decomposition. We discuss various programmatic issues and provide a set of design guidelines for GPU programming that are instructive to avoid common pitfalls as well as to extract performance from the GPU architecture. Conclusions: We demonstrate that GPU algorithms represent a significant technological advance for the simulation of complex biological models. We further demonstrate with our epidermal model that the integration of multiple complex modeling methods for heterogeneous multicellular biological processes is both feasible and computationally tractable using this new technology. We hope that the provided algorithms and source code will be a starting point for modelers to develop their own GPU implementations, and encourage others to implement their modeling methods on the GPU and to make that code available to the wider community.

show abstract

“…However, GPUs have limited instructions and limited parallelism relative to FPGA's configurability. The research in [10] employed acceleration using GPU. Another approach uses clusters of workstations [12].…”

Section: Introductionmentioning

confidence: 99%