Design of the GraphBLAS API for C

Buluç, Aydın; Mattson, Tim; McMillan, Scott; Moreira, José E.; Yang, Carl

doi:10.1109/ipdpsw.2017.117

Cited by 73 publications

(57 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Instead, we touch on the computational building blocks for the machine learning algorithms that are commonly applied to genomic and proteomic data. A large class of machine learning methods are built on top of basic linear algebraic subroutines that are found in the modern dense BLAS [51], Sparse BLAS [52], or the GraphBLAS [53]. This relationship is illustrated in Figure 3.…”

Section: Machine Learning For Genomics and Proteomicsmentioning

confidence: 99%

The parallelism motifs of genomic data analysis

Yelick

Buluç

Awan

et al. 2020

Phil. Trans. R. Soc. A.

View full text Add to dashboard Cite

Genomic datasets are growing dramatically as the cost of sequencing continues to decline and small sequencing devices become available. Enormous community databases store and share these data with the research community, but some of these genomic data analysis problems require large-scale computational platforms to meet both the memory and computational requirements. These applications differ from scientific simulations that dominate the workload on high-end parallel systems today and place different requirements on programming support, software libraries and parallel architectural design. For example, they involve irregular communication patterns such as asynchronous updates to shared data structures. We consider several problems in high-performance genomics analysis, including alignment, profiling, clustering and assembly for both single genomes and metagenomes. We identify some of the common computational patterns or ‘motifs’ that help inform parallelization strategies and compare our motifs to some of the established lists, arguing that at least two key patterns, sorting and hashing, are missing. This article is part of a discussion meeting issue ‘Numerical algorithms for high-performance computational science’.

show abstract

Section: Machine Learning For Genomics and Proteomicsmentioning

confidence: 99%

The parallelism motifs of genomic data analysis

Yelick

Buluç

Awan

et al. 2020

Phil. Trans. R. Soc. A.

View full text Add to dashboard Cite

show abstract

“…Multiplication on a semiring allows the user to overload scalar multiplication and addition operations and still use the same SpGEMM algorithm. Many existing SpGEMM implementations support user-defined semirings, including those that implement the GraphBLAS API (Buluç et al, 2017).…”

Section: Sparse Matrix Construction and Multiplicationmentioning

confidence: 99%

BELLA: Berkeley Efficient Long-Read to Long-Read Aligner and Overlapper

Guidi

Ellis

Rokhsar

et al. 2018

Preprint

View full text Add to dashboard Cite

Recent advances in long-read sequencing enable the characterization of genome structure and its intra-and inter-species variation at a resolution that was previously impossible. Detecting overlaps between reads is integral to many long-read genomics pipelines, such as de novo genome assembly. While longer reads simplify genome assembly and improve the contiguity of the reconstruction, current long-read technologies come with high error rates. We present Berkeley Long-Read to Long-Read Aligner and Overlapper (BELLA), a novel algorithm for computing overlaps and alignments via sparse matrix-matrix multiplication that balances the goals of recall and precision, performing well on both.We present a probabilistic model that demonstrates the feasibility of using short k-mers for detecting candidate overlaps. We then introduce a notion of reliable k-mers based on our probabilistic model. Combining reliable k-mers with our binning mechanism eliminates both the k-mer set explosion that would otherwise occur with highly erroneous reads, and the spurious overlaps from k-mers originating in repetitive regions. Finally, we present a new method based on Chernoff bounds for separating true overlaps from false positives using a combination of alignment techniques and probabilistic modeling. Our methodologies aim at maximizing the balance between precision and recall. On both real and synthetic data, BELLA performs amongst the best in terms of F1 score, showing a performance stability which is often missing for competitor software. BELLA's F1 score is consistently within 1.7% of the top entry. Notably, we show improved de novo assembly results on synthetic data when coupling BELLA with the Miniasm assembler.Long-read technologies (Eid et al., 2009;Goodwin et al., 2015) generate long reads with average lengths reaching and often exceeding 10,000 base pairs (bp). These allow the resolution of complex genomic repetitions, enabling more accurate ensemble views that were not possible with previous short-read technologies (Phillippy et al., 2008;Nagarajan and Pop, 2009). However, the improved read length of these technologies comes at the cost of lower accuracy, with error rates ranging from 5% to 35%. Nevertheless, errors are more random and more evenly distributed within Pacific Biosciences long-read data (Giordano et al., 2017) compared to short-read technologies.The majority of the state-of-the-art long-read assemblers uses the Overlap-Layout-Consensus (OLC) paradigm (Berlin et al., 2015). The first step in OLC assembly consists of detecting overlaps between reads to construct an overlap (or string) graph. The OLC paradigm benefits from longer reads as significantly fewer reads are required to cover the genome, limiting the size of the overlap graph. Highly-accurate overlap detection is a major computational bottleneck in OLC assembly (Myers, 2014), mainly due to the compute-intensive nature of pairwise alignment.At present, several algorithms are capable of overlapping error-prone long-read data with varying accuracy. The prevaili...

show abstract

“…This section gives a brief overview of important elements of Version 1.0 of the GraphBLAS C API specification [2] in order give some context for future extensions proposed in this paper. The full specification can be downloaded from the GraphBLAS Forum website [1].…”

Section: The Graphblas C Apimentioning

confidence: 99%

“…The GraphBLAS aims to standardize the mathematical concepts [5] and the application programming interface (API) [2] for performing graph computations in the language of linear algebra [6]. The GraphBLAS C API version 1.0 has been provisionally released in May 2017 [3].…”

Section: Introductionmentioning

confidence: 99%

GraphBLAS C API: Ideas for future versions of the specification

Mattson

Yang

McMillan

et al. 2017

2017 IEEE High Performance Extreme Computing Conference (HPEC)

Self Cite

View full text Add to dashboard Cite

Abstract-The GraphBLAS C specification provisional release 1.0 is complete. To manage the scope of the project, we had to defer important functionality to a future version of the specification. For example, we are well aware that many algorithms benefit from an inspector-executor execution strategy. We also know that users would benefit from a number of standard predefined semirings as well as more general user-defined types. These and other features are described in this paper in the context of a future release of the GraphBLAS C API.

show abstract

Design of the GraphBLAS API for C

Cited by 73 publications

References 12 publications

The parallelism motifs of genomic data analysis

The parallelism motifs of genomic data analysis

BELLA: Berkeley Efficient Long-Read to Long-Read Aligner and Overlapper

GraphBLAS C API: Ideas for future versions of the specification

Contact Info

Product

Resources

About