项目名称: 下一代测序数据自适应错误修正技术的研究
项目编号: No.61472082
项目类型: 面上项目
立项/批准年度: 2015
项目学科: 计算机科学学科
项目作者: 林劼
作者单位: 福建师范大学
项目金额: 63万元
中文摘要: 在下一代测序技术中,错误修正模型是序列拼接的基础,是正确有效测序的重要保证,也是近年生物信息学研究的热点之一。本课题拟通过研究下一代测序数据特征及其错误的分布特点,建立数据质量模型,为错误修正技术提供数据自适应模型。本课题计划对测序数据进行高覆盖细粒度分组聚合,将相似的测序数据聚合在同一组内,应用错误判别模型识别组内错误数据,并进行组内错误修正处理。为了有效利用有限的计算资源来处理海量数据,本课题将采用分布式计算框架,从而达到快速高效的错误修正目的,为测序技术的实际应用提供支持。本课题的研究成果可以结合目前高速发展的下一代测序技术应用在生物科学中的研究和临床疾病的检测,如个性化医疗等领域。
中文关键词: 生物信息处理;下一代测序技术;错误修正;序列分析;聚合模型
英文摘要: In next-generation sequencing(NGS), error correction in short reads is critical in assembly of high quality sequences.In this proposal, we propose to study the characteristic of short reads data generated from NGS and its associated errors,build appropriate quality models to guide error correction process. We will investigate cluster models which have high coverage and fine grain ability to group similar and neigborhood short reads into a cluster. Then errors are corrected in individual cluster separately which is distributed in different computing nodes. In order to utilize limited computing facility to cope with large-scale volume of NGS data, we will study distributed framework which will speed up the computing time, decrease the requirement memory usage,and result in more acurate short reads for assemble. The proposed research will benefit high throughput NGS applications both in research and in practice,ie. personalized medicine.
英文关键词: bioinformaics;NGS;error correction;sequence analysis;cluster model