arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2211.03079cs.ARcs.DCq-bio.GN
收藏

RUBICON:一种用于设计高效的基于深度学习的基因组碱基识别工具的框架

RUBICON: A Framework for Designing Efficient Deep Learning-Based Genomic Basecallers

Gagandeep Singh, Mohammed Alser, Kristof Denolf, Can Firtina, Alireza Khodamoradi, Meryem Banu Cavlak, Henk Corporaal, Onur Mutlu

AI总结:

针对现有基于深度学习的基因组碱基识别工具计算与内存成本高的问题,提出包含QABAS框架和SkipClip技术的RUBICON框架,开发出的硬件优化碱基识别工具RUBICALL在速度和精度上均优于现有最优工具。

摘要:

纳米孔测序会产生含噪声的电信号,需要通过名为碱基识别(basecalling)的计算步骤将其转换为标准的DNA核苷酸碱基序列。碱基识别的准确性和速度对基因组分析的所有后续步骤都有至关重要的影响。许多研究人员采用复杂的基于深度学习的模型来执行碱基识别,却没有考虑这类模型的计算需求,导致碱基识别工具速度慢、效率低且内存消耗大。因此,需要在保持准确性的同时降低碱基识别的计算和内存成本。我们的目标是开发一个综合性框架,用于构建兼具高效率和高性能的基于深度学习的碱基识别工具。我们提出了RUBICON,这是一款用于开发硬件优化型碱基识别工具的框架。RUBICON包含两种专门为碱基识别设计的新型机器学习技术。首先,我们提出了首个量化感知碱基识别神经架构搜索(QABAS)框架,该框架可针对给定的硬件加速平台定制碱基识别神经网络架构,同时共同探索并找到每个神经网络层的最佳位宽精度。其次,我们开发了SkipClip,这是首个用于移除现代碱基识别工具中存在的跳跃连接的技术,可在不损失任何碱基识别准确性的前提下大幅降低资源和存储需求。我们通过开发RUBICALL(首款可实现快速准确碱基识别的硬件优化型碱基识别工具)证明了RUBICON的优势。与最快的现有先进碱基识别工具相比,RUBICALL实现了3.96倍的速度提升,同时准确性提高了2.97%。我们的研究表明,RUBICON可帮助研究人员开发出优于专家设计模型的硬件优化型碱基识别工具。

英文摘要:

Nanopore sequencing generates noisy electrical signals that need to be converted into a standard string of DNA nucleotide bases using a computational step called basecalling. The accuracy and speed of basecalling have critical implications for all later steps in genome analysis. Many researchers adopt complex deep learning-based models to perform basecalling without considering the compute demands of such models, which leads to slow, inefficient, and memory-hungry basecallers. Therefore, there is a need to reduce the computation and memory cost of basecalling while maintaining accuracy. Our goal is to develop a comprehensive framework for creating deep learning-based basecallers that provide high efficiency and performance. We introduce RUBICON, a framework to develop hardware-optimized basecallers. RUBICON consists of two novel machine-learning techniques that are specifically designed for basecalling. First, we introduce the first quantization-aware basecalling neural architecture search (QABAS) framework to specialize the basecalling neural network architecture for a given hardware acceleration platform while jointly exploring and finding the best bit-width precision for each neural network layer. Second, we develop SkipClip, the first technique to remove the skip connections present in modern basecallers to greatly reduce resource and storage requirements without any loss in basecalling accuracy. We demonstrate the benefits of RUBICON by developing RUBICALL, the first hardware-optimized basecaller that performs fast and accurate basecalling. Compared to the fastest state-of-the-art basecaller, RUBICALL provides a 3.96x speedup with 2.97% higher accuracy. We show that RUBICON helps researchers develop hardware-optimized basecallers that are superior to expert-designed models.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑