Big Data Proteogenomics lies at the intersection of high-throughput Mass Spectrometry (MS) based proteomics and Next Generation Sequencing based genomics. The combined and integrated analysis of these two high-throughput technologies can help discover novel proteins using genomic, and transcriptomic data. Due to the biological significance of integrated analysis, the recent past has seen an influx of proteogenomic tools that perform various tasks, including mapping proteins to the genomic data, searching experimental MS spectra against a six-frame translation genome database, and automating the process of annotating genome sequences. To date, most of such tools have not focused on scalability issues that are inherent in proteogenomic data analysis where the size of the database is much larger than a typical protein database. These state-of-the-art tools can take more than half a month to process a small-scale dataset of one million spectra against a genome of 3 GB. In this article, we provide an up-to-date review of tools that can analyze proteogenomic datasets, providing a critical analysis of the techniques' relative merits and potential pitfalls. We also point out potential bottlenecks and recommendations that can be incorporated in the future design of these workflows to ensure scalability with the increasing size of proteogenomic data. Lastly, we make a case of how high-performance computing (HPC) solutions may be the best bet to ensure the scalability of future big data proteogenomic data analysis.
Very fast transient overvoltages (VFTO) generated in a gas insulated substation (GIS) during disconnector switching operations is one of the major concerns in insulation design of the GIS. Such over voltages can cause malfunctioning of the protection and control circuits in addition to initiating faults inside the gas insulated bus ducts of the substation especially in the presence of metallic particles. In this paper, VFTO have been estimated at various points in a 1000 kV rated substation for all possible valid disconnector switching operations. For the substation studied, the maximum computed overvoltage is 1.58 pu without considering the trapped charge on the busbar. Major frequency components in the simulated VFTO lies in the range of 840 KHz to 30 MHz. Simulations have been carried out using the Electromagnetic Transient Program (EMTP).
Effective programming models offer programmers the ability to harness the capabilities of the underlying platform. For decades, the two-sided Message Passing Interface (MPI) has become a de facto standard for communication among processes running on distributed memory systems. As high-performance GPU computing becomes the trend, GPU-initiated one-sided communication becomes a viable solution for multi-GPU scaling. It also highlights the use of one-sided communication on CPUs. However, the lack of deep understanding of one-sided communication performance and its impact on an application’s performance becomes a hurdle. In this paper, we overcome this hurdle by proposing a Message Roofline model, which characterizes an application’s sustained messaging performance (GB/s) as a function of its message size, number of messages per synchronization, peak network bandwidth, and network latency. We use three benchmarks to demonstrate the potentials of one-sided communication on CPUs and GPUs. These benchmarks include Stencils representing applications that follow the bulk synchronization programming model, Sparse Triangular Solve representing directed acyclic graph computations that perform asynchronous point-to-point communications, and Distributed HashTable performing atomic compare and swaps. Our evaluation provides insights into practically understanding the two-sided and one-sided communications in MPI applications, and can also guide hardware vendors with design principles lest the potential performance of one-sided communications being under-utilized.
The most commonly employed method for peptide identification in mass-spectrometry based proteomics involves comparing experimentally obtained tandem MS/MS spectra against a set of theoretical MS/MS spectra. The theoretical MS/MS spectra data are predicted using protein sequence database. Most state-of-the-art peptide search algorithms index theoretical spectra data to quickly filter-in the relevant (similar) indexed spectra when searching an experimental MS/MS spectrum. Data filtration substantially reduces the required number of computationally expensive spectrum-to-spectrum comparison operations. However, the number of predicted (and indexed) theoretical spectra grows exponentially with increase in post-translational modifications creating a memory and I/O bottleneck. In this paper, we present a parallel algorithm, called LBE, for efficient partitioning of theoretical spectra data on a distributed-memory architecture. Our proposed algorithm first groups the similar theoretical spectra. The groups are then finely split across the system allowing machines to perform almost equal amount of work when querying a MS/MS spectrum. Our results show that the compute load imbalance using LBE based data distribution is ≤ 20% allowing speedups of order of magnitudes over existing methods. The proposed algorithm has been implemented on a compute cluster using MPI library. Experimental results for increasing index sizes are reported in terms of execution time, speedups and memory footprint. To the best of our knowledge, LBE is the first load-balancing technique for MS/MS proteomics data on memory-distributed clusters that incorporates proteomics domain knowledge for efficient load-balancing. Source code is made available at: https://github.com/pcdslab/lbdslim/tree/mpi.
17 November 2021 in Journal of Parallel and Distributed Computing