Recovering genomes from shotgun metagenomic sequence data allows detailed taxonomic and functional characterization of individual species or strains in a microbial community. Retrieving these metagenome-assembled genomes (MAGs) involves seven stages. First, low-quality bases, along with adapter and host sequences, are removed. Second, overlapping sequences are assembled to create longer contiguous fragments. Third, these fragments are clustered based on sequence composition and abundance. Fourth, these sequence clusters, or bins, undergo rounds of quality assessment and refinement to yield MAGs. The optional fifth stage is dereplication of MAGs to select representatives. Next, each MAG is taxonomically classified. The optional seventh stage is assessing the fraction of diversity that has been recovered. The output of this protocol is draft genomes, which can provide invaluable clues about uncultured organisms. This protocol takes ~1 week to run, depending on computational resources available, and requires prior experience with high-performance computing, shell script programming and Python.

从鸟枪法宏基因组序列数据中恢复基因组可以对微生物群落中的单个物种或菌株进行详细的分类学和功能表征。检索这些宏基因组组装的基因组 (MAG) 涉及七个阶段。首先，去除低质量的碱基以及衔接子和宿主序列。其次，组装重叠序列以创建更长的连续片段。第三，这些片段根据序列组成和丰度进行聚类。第四，这些序列簇或箱经过多轮质量评估和改进以产生 MAG。可选的第五阶段是 MAG 去复制以选择代表。接下来，对每个 MAG 进行分类。可选的第七阶段是评估已恢复的多样性部分。该协议的输出是草图基因组，这可以提供有关未培养生物的宝贵线索。该协议需要大约 1 周的时间运行，具体取决于可用的计算资源，并且需要具有高性能计算、shell 脚本编程和 Python 的先前经验。