本数据库提供了经纬度等环境信息的参考，这些信息来源于公开的基因组数据。这些公开可用的基因组数据是从 Figshare（https://doi.org/10.6084/m9.figshare.4902923）下载的，涉及 Tara Oceans 的手动整理的 MAGs、从 ENA 下载了项目编号为 PRJEB33281 的 GORG 数据、以及从 https://mmp2.sfb.uit.no/databases/ 下载的 MarDB 数据。GEM MAGs 是从 https://portal.nersc.gov/GEM/ 下载的。GTDB r89 中包含的 MAGs 是从 https://data.gtdb.ecogenomic.org/releases/release89/ 下载的。MIBiG 和 BiG-FAM 数据库分别可以通过 https://mibig.secondarymetabolites.org/ 和 https://bigfam.bioinformatics.nl/ 访问。所有原始的宏转录组测序读数可以通过 ENA 访问，网址为 https://www.ebi.ac.uk/ena，使用的标识符列在 https://doi.org/10.5281/zenodo.3473199。我们要感谢那些负责生成宏基因组数据（Tara Oceans、Malaspina，以及 Biogeotraces、HOT 和 BATS）以及制作公开可用基因组（GORG SAGs、MAR 数据库和手动整理的 MAGs）的个人和团队。

Publicly available genomes were downloaded from Figshare (https://doi.org/10.6084/m9.figshare.4902923) for manually curated MAGs from Tara Oceans, from ENA using project accession PRJEB33281 for GORG, and from https://mmp2.sfb.uit.no/databases/ for MarDB. The GEM MAGs were downloaded from https://portal.nersc.gov/GEM/. MAGs included in GTDB r89 were downloaded from https://data.gtdb.ecogenomic.org/releases/release89/. The MIBiG and BiG-FAM databases can be accessed at https://mibig.secondarymetabolites.org/ and https://bigfam.bioinformatics.nl/, respectively. All raw metatranscriptomic reads are available via ENA at https://www.ebi.ac.uk/ena using the identifiers listed at https://doi.org/10.5281/zenodo.3473199. We would like to express our gratitude to the individuals and teams responsible for generating the metagenomic data (Tara Oceans, Malaspina, as well as Biogeotraces, HOT, and BATS) and for producing the publicly available genomes (GORG SAGs, The MAR Databases, and manually curated MAGs).

使用R包“Hmsic”计算了8308个海洋微生物基因组中不同代谢功能（KEGG模块）定性矩阵的Pearson相关性，并保留了相关系数大于0.6且P值小于0.05的数据（https://cran.r-project.org/ ）。然后，我们利用交互式平台Gephi 0.10.1（https://gephi.org/ ）中的Louvain算法（https://github.com/topics/louv-algorithm ）来识别代谢策略。Louvain算法是一种层次聚类算法，它递归地将社区合并为一个单一节点，并在凝聚图上执行模块化聚类。连接每个组件中节点的关系具有属性权重，该权重决定了关系的强度。通过这种方法识别出了海洋微生物八种主要的代谢策略。

We calculated the Pearson’s correlation of different metabolic function (KEGG module) qualitative matrix in 9504 marine microbial genomes using the R-package “Hmsic,” and retained data with correlation coefficient > 0.6 and P < 0.05 (https://cran.r-project.org/). Then we used Louvain algorithm (https://github.com/topics/louvain-algorithm) from the interactive platform Gephi 0.10.1 (https://gephi.org/) to identify metabolic strategy. The Louvain algorithm is a hierarchical clustering algorithm, that recursively merges communities into a single node and executes the modularity clustering on the condensed graphs. The relationships that connect the nodes in each component have a property weight which determines the strength of the relationship.

Category of KEGG Module

KEGG Module的注释

收集的基因组编号

The design of this database is based on the metabolic strategy of Marine microorganisms to quickly match and query the appropriate culture conditions by entering the microbial classification name provided by GTDB (Genome Taxonomy Database). The database uses the information of 8308 Marine microbial genomes for metabolic strategy analysis, referring to the culture conditions of 16932 strains of Marine microbial culture, and has been verified by the culture of new strains.