Methods for Evolutionary Genomics Analysis

项目摘要 此管理补充请求旨在开发支持云的高度可扩展版本的 分子进化遗传学分析软件（MEGA-CC： www.megasoftware.net）。MEGA-CC的开发是NIH资助的 研究项目，开发机器学习方法和工具，用于分子生物学的比较分析。 序列的 随着基因组测序的巨大进步，研究人员正在组装包含大量 物种、菌株、基因和基因组片段。这些数据的系统基因组分析对于 了解病原体，人类和物种在生命之树中的进化变化动态。 用于基因组学的机器学习方法和软件工具现在是必要的，因为 基因组数据集的大小限制了目前可用的方法和工具的实际效用， 过多的计算时间和内存需求。资助赠款的一个组成部分是执行 我们在MEGA软件套件（www.megasoftware.net）中的新机器学习方法， 流行的生物信息学软件（年内超过20，000篇同行评审引文和350，000次软件下载 仅2020年）。MEGA软件包括用于组装序列比对的大量工具， 推断进化树，估计遗传距离和差异，推断祖先序列， 计算时间树和测试选择。现在，所有研究调查都要求进行这些分析， 使用多个DNA或RNA序列的领域。 然而，MEGA及其计算核心（MEGA-CC）并没有针对在 云基础设施和高性能计算集群。这笔补助金将使 我们将推进MEGA for Cloud Ready，以利用可扩展性、弹性计算能力和 通过云基础设施（MEGA-CR）实现软件升级和维护。它也将使MEGA 可与现有和未来的云基础设施互操作。此外，此补充将有助于使用 MEGA中的新机器学习方法与实践中的大基因组数据，从而解决了迫在眉睫的 快速增长的需要越来越大的社区研究人员使用MEGA。MEGA-CR将增加 科学界分析大型数据集的MEGA可用性， 成本效益和云就绪性的可扩展性变得至关重要。

PROJECT SUMMARY This administrative supplement request aims to develop a cloud-enabled, highly scalable version of the computational core of the Molecular Evolutionary Genetics Analysis software (MEGA-CC: www.megasoftware.net). The development of MEGA-CC is a significant component of the NIH-funded research project to develop machine learning methods and tools for comparative analysis of molecular sequences. With big advances in genome sequencing, researchers are assembling datasets containing large numbers of species, strains, genes, and genomic segments. Phylogenomic analyses of these data are essential to understanding the dynamics of evolutionary change of pathogens, humans, and species across the tree of life. Machine learning methods and software tools for phylogenomics are now necessary because the expanding size of phylogenomic datasets limits the practical utility of currently available methods and tools due to excessive computational time and memory requirements. One component of the funded grant is implementing our new machine learning methods in the MEGA software suite (www.megasoftware.net), an extremely popular bioinformatics software (>20,000 peer-reviewed citations and 350,000 software downloads in the year 2020 alone). The MEGA software includes a large repertoire of tools for assembling sequence alignments, inferring evolutionary trees, estimating genetic distances and diversities, inferring ancestral sequences, computing timetrees, and testing selection. These analyses are now required in all research investigations and fields in which multiple DNA or RNA sequences are used. However, MEGA and its computational core (MEGA-CC) are not optimized for distribution and execution on cloud infrastructure and high-performance computing clusters. This supplement to the funded grant will enable us to advance MEGA for cloud readiness to harness the scalability, elastic computing power, and easy software upgrade and maintenance enabled by cloud infrastructure (MEGA-CR). It will also make MEGA interoperable with existing and future cloud infrastructure. Additionally, this supplement will facilitate using the new machine learning methods in MEGA with big genomic data in practice, thus addressing an imminent and fast-growing need for an increasingly larger community of researchers using MEGA. MEGA-CR will increase the usability of MEGA for the scientific community analyzing very large datasets for which greater accessibility, cost-efficiency, and scalability of cloud-readiness is becoming crucial.