ABSTRACT Whole-genome sequencing has boosted our ability to explore microbial diversity by enabling the recovery of metagenome-assembled genomes (MAGs) directly from environmental DNA. As a result, the vast availability of sequencing data has prompted the development of numerous bioinformatics pipelines for MAG reconstruction, along with challenges to identify the most suitable pipeline to perform the analysis according to the user needs. This report briefly discusses the computational requirements of these pipelines; presents the variety of interfaces, workflow managers, and package managers they feature; and describes the typical modular structure. Also, it provides a compacted technical overview of 41 publicly available pipelines or platforms to build MAGs starting from short and/or long sequences. Moreover, recognizing the overwhelming number of factors to consider when selecting an appropriate pipeline, we introduce an interactive decision-support web application, 2Pipe, that helps users to identify a suitable workflow based on their input data characteristics, desired outcomes, and computational constraints. The tool presents a question-driven interface to customize the recommendation, a pipeline gallery to offer a summarized description, and a pipeline comparison based on key factors used for the questionnaire. Beyond this and foreseeing the release of novel pipelines in the near future, we include a quick form and detailed instructions for developers to append their workflow in the application. Altogether, this review and the application equip the researchers with a general outlook of the growing metagenomics pipeline landscape and guide the users toward deciding the workflow that best fits their expectations and infrastructure.
Yepes-García et al. (Thu,) studied this question.