An Out-of-Core GPU based Dimensionality Reduction Algorithm for Big Mass Spectrometry Data and Its Application in Bottom-up Proteomics

Awan, Muaaz Gul; Saeed, Fahad

doi:10.1145/3107411.3107466

Cited by 10 publications

(3 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To handle large matrices by overcoming memoryconstraints of a single-node or a GPU, out-of core computations [51], [1], [7], [9] have been used which seek to decompose a matrix into smaller pieces and operate on them in multiple passes; data resides in disk and has to be explicitly moved in and out of memory for the passes. While this addresses the problem of handling large matrices on memory-constrained single-nodes and accelerators, the achievable speedup is constrained by the sequential passes that the algorithm has to make.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Extending Shared-Memory Computations to Multiple Distributed Nodes

Ahmed¹

2020

IJACSA

View full text Add to dashboard Cite

With the emergence of accelerators like GPUs, MICs and FPGAs, the availability of domain specific libraries (like MKL) and the ease of parallelization associated with CUDA and OpenMP based shared-memory programming, node-based parallelization has recently become a popular choice among developers in the field of scientific computing. This is evident from the large volume of recently published work in various domains of scientific computing, where shared-memory programming and accelerators have been used to accelerate applications. Although these approaches are suitable for small problem-sizes, there are issues that need to be addressed for them to be applicable to larger input domains. Firstly, the primary focus of these works has been to accelerate the core kernel; acceleration of input/output operations is seldom considered. Many operations in scientific computing operate on large matrices-both sparse and dense-that are read from and written to external files. These input-output operations present themselves as bottlenecks and significantly effect the overall application time. Secondly, node-based parallelization limits a developer from distributing the computation beyond a single node without him having to learn an additional programming paradigm like MPI. Thirdly, the problem size that can be effectively handled by a node is limited by the memory of the node and accelerator. In this paper, an Asynchronous Multi-node Execution (AMNE) approach is presented that uses a unique combination of the shared-file system and pseudo-replication to extend node-based algorithms to a distributed multiple node implementation with minimal changes to the original node-based code. We demonstrate this approach by applying it to GEMM, a popular kernel in dense linear algebra and show that the presented methodology significantly advances the state of art in the field of parallelization and scientific computing.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Many applications in engineering and scientific computing involve operations on dense and sparse matrices [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11]. At the core of many of these applications is the General Matrix-Matrix multiplication (GEMM) operation, which is regarded as one of the most widely used high performance kernels.…”

Section: Introductionmentioning

confidence: 99%

Extending Shared-Memory Computations to Multiple Distributed Nodes

Ahmed¹

2020

IJACSA

View full text Add to dashboard Cite

show abstract

“…Many algorithms based on parallel computing techniques have been proposed in different fields like genomics, proteogenomics, clinical informatics, imaging informatics, etc. [ 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 ]. Using GPUs for accelerating these type of problems has become very popular recently.…”

Section: Introductionmentioning

confidence: 99%

Fast-GPU-PCC: A GPU-Based Technique to Compute Pairwise Pearson’s Correlation Coefficients for Time Series Data—fMRI Study

Eslami

Saeed

2018

High-Throughput

Self Cite

View full text Add to dashboard Cite

Functional magnetic resonance imaging (fMRI) is a non-invasive brain imaging technique, which has been regularly used for studying brain’s functional activities in the past few years. A very well-used measure for capturing functional associations in brain is Pearson’s correlation coefficient. Pearson’s correlation is widely used for constructing functional network and studying dynamic functional connectivity of the brain. These are useful measures for understanding the effects of brain disorders on connectivities among brain regions. The fMRI scanners produce huge number of voxels and using traditional central processing unit (CPU)-based techniques for computing pairwise correlations is very time consuming especially when large number of subjects are being studied. In this paper, we propose a graphics processing unit (GPU)-based algorithm called Fast-GPU-PCC for computing pairwise Pearson’s correlation coefficient. Based on the symmetric property of Pearson’s correlation, this approach returns Nfalse(N−1false)/2 correlation coefficients located at strictly upper triangle part of the correlation matrix. Storing correlations in a one-dimensional array with the order as proposed in this paper is useful for further usage. Our experiments on real and synthetic fMRI data for different number of voxels and varying length of time series show that the proposed approach outperformed state of the art GPU-based techniques as well as the sequential CPU-based versions. We show that Fast-GPU-PCC runs 62 times faster than CPU-based version and about 2 to 3 times faster than two other state of the art GPU-based methods.

show abstract

MaSS‐Simulator: A Highly Configurable Simulator for Generating MS/MS Datasets for Benchmarking of Proteomics Algorithms

Awan

Saeed

2018

Proteomics

Self Cite

View full text Add to dashboard Cite

Mass Spectrometry (MS)-based proteomics has become an essential tool in the study of proteins. With the advent of modern MS machines huge amounts of data is being generated, which can only be processed by novel algorithmic tools. However, in the absence of data benchmarks and ground truth datasets algorithmic integrity testing and reproducibility is a challenging problem. To this end, MaSS-Simulator has been presented, which is an easy to use simulator and can be configured to simulate MS/MS datasets for a wide variety of conditions with known ground truths. MaSS-Simulator offers many configuration options to allow the user a great degree of control over the test datasets, which can enable rigorous and large- scale testing of any proteomics algorithm. MaSS-Simulator is assessed by comparing its performance against experimentally generated spectra and spectra obtained from NIST collections of spectral library. The results show that MaSS-Simulator generated spectra match closely with real-spectra and have a relative-error distribution centered around 25%. In contrast, the theoretical spectra for same peptides have relative-error distribution centered around 150%. MaSS-Simulator will enable developers to specifically highlight the capabilities of their algorithms and provide a strong proof of any pitfalls they might face. Source code, executables, and a user manual for MaSS-Simulator can be downloaded from https://github.com/pcdslab/MaSS-Simulator.

show abstract

An Out-of-Core GPU based Dimensionality Reduction Algorithm for Big Mass Spectrometry Data and Its Application in Bottom-up Proteomics

Cited by 10 publications

References 20 publications

Extending Shared-Memory Computations to Multiple Distributed Nodes

Extending Shared-Memory Computations to Multiple Distributed Nodes

Fast-GPU-PCC: A GPU-Based Technique to Compute Pairwise Pearson’s Correlation Coefficients for Time Series Data—fMRI Study

MaSS‐Simulator: A Highly Configurable Simulator for Generating MS/MS Datasets for Benchmarking of Proteomics Algorithms

Contact Info

Product

Resources

About