利用解旋酶隐马尔可夫模型对纳米孔测序过程进行建模
Modelling the nanopore sequencing process with Helicase HMMs
AI总结:
本文提出解旋酶隐马尔可夫模型(HHMM),将解旋酶马达蛋白动力学与核苷酸序列结合建模,支持对数百万状态的分析,为高分辨率纳米孔测序的碱基识别算法开发提供了新基础。
摘要:
纳米孔测序技术的最新进展,特别是Oxford Nanopore Technology的R10纳米孔,要求开发改进的数据处理方法,以充分利用其超过9-mer分辨率的潜力。离子电流的处理主要采用基于神经网络的方法,这些方法以高碱基识别准确率著称,但在更高分辨率下面临发展瓶颈。鉴于此,我们提出了解旋酶隐马尔可夫模型(HHMM),这是一个新颖的框架,旨在将解旋酶马达蛋白的动力学与纳米孔测序过程中的核苷酸序列结合起来进行建模。该模型支持对数百万个不同状态的分析,增强了我们对原始离子电流及其与核苷酸序列比对的理解。我们的研究结果表明,HHMM不仅是一种有效的可视化工具,而且是开发先进碱基识别算法的有效基础。该方法为充分利用新兴高分辨率纳米孔测序技术的全部能力提供了一条有前景的途径。
英文摘要:
Recent advancements in nanopore sequencing technology, particularly the R10 nanopore from Oxford Nanopore Technology, have necessitated the development of improved data processing methods to utilize their potential for more than 9-mer resolution fully. The processing of the ion currents predominantly utilizes neural network-based methods known for their high basecalling accuracy but face developmental bottlenecks at higher resolutions. In light of this, we introduce the Helicase Hidden Markov Model (HHMM), a novel framework designed to incorporate the dynamics of the helicase motor protein alongside the nucleotide sequence during nanopore sequencing. This model supports the analysis of millions of distinct states, enhancing our understanding of raw ion currents and their alignment with nucleotide sequences. Our findings demonstrate the utility of HHMM not only as a potent visualization tool but also as an effective base for developing advanced basecalling algorithms. This approach offers a promising avenue for leveraging the full capabilities of emerging high-resolution nanopore sequencing technologies.
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。