NGScloud2: optimized bioinformatic analysis using Amazon Web Services

PeerJ

article

Fernando Mora-Márquez¹  José Luis Vázquez-Poletti²  Unai López de Heredia¹ 

[1] GI Sistemas Naturales e Historia Forestal, Dpto. Sistemas y Recursos Naturales, ETSI Montes, Forestal y del Medio Natural, Universidad Politécnica de Madrid;GI Arquitectura de Sistemas Distribuidos, Dpto. de Arquitectura de Ordenadores y Automática, Facultad de Informática, Universidad Complutense de Madrid

关键词: AWS;    Bioinformatics;    Cloud computing;    Functional annotation;    Next generation sequencing;    Transcriptomics;   

DOI  :  10.7717/peerj.11237

学科分类：社会科学、人文和艺术（综合）

来源: Inra

【 摘 要 】

BackgroundNGScloud was a bioinformatic system developed to perform de novo RNAseq analysis of non-model species by exploiting the cloud computing capabilities of Amazon Web Services. The rapid changes undergone in the way this cloud computing service operates, along with the continuous release of novel bioinformatic applications to analyze next generation sequencing data, have made the software obsolete. NGScloud2 is an enhanced and expanded version of NGScloud that permits the access to ad hoc cloud computing infrastructure, scaled according to the complexity of each experiment.MethodsNGScloud2 presents major technical improvements, such as the possibility of running spot instances and the most updated AWS instances types, that can lead to significant cost savings. As compared to its initial implementation, this improved version updates and includes common applications for de novo RNAseq analysis, and incorporates tools to operate workflows of bioinformatic analysis of reference-based RNAseq, RADseq and functional annotation. NGScloud2 optimizes the access to Amazon’s large computing infrastructures to easily run popular bioinformatic software applications, otherwise inaccessible to non-specialized users lacking suitable hardware infrastructures.ResultsThe correct performance of the pipelines for de novo RNAseq, reference-based RNAseq, RADseq and functional annotation was tested with real experimental data, providing workflow performance estimates and tips to make optimal use of NGScloud2. Further, we provide a qualitative comparison of NGScloud2 vs. the Galaxy framework. NGScloud2 code, instructions for software installation and use are available at https://github.com/GGFHF/NGScloud2. NGScloud2 includes a companion package, NGShelper that contains Python utilities to post-process the output of the pipelines for downstream analysis at https://github.com/GGFHF/NGShelper.

【 授权许可】

CC BY