DRAM for distilling microbial metabolism to automate the curation of microbiome function

Abstract

Microbial and viral communities transform the chemistry of Earth's ecosystems, yet the specific reactions catalyzed by these biological engines are hard to decode due to the absence of a scalable, metabolically resolved, annotation software. Here, we present DRAM (Distilled and Refined Annotation of Metabolism), a framework to translate the deluge of microbiome-based genomic information into a catalog of microbial traits. To demonstrate the applicability of DRAM across metabolically diverse genomes, we evaluated DRAM performance on a defined, in silico soil community and previously published human gut metagenomes. We show that DRAM accurately assigned microbial contributions to geochemical cycles and automated the partitioning of gut microbial carbohydrate metabolism at substrate levels. DRAM-v, the viral mode of DRAM, established rules to identify virally-encoded auxiliary metabolic genes (AMGs), resulting in the metabolic categorization of thousands of putative AMGs from soils and guts. Together DRAM and DRAM-v provide critical metabolic profiling capabilities that decipher mechanisms underpinning microbiome function.

MicrobeAnnotator: a user-friendly, comprehensive functional annotation pipeline for microbial genomes

“…The first comparison was in terms of database entries used by each tool. As previously suggested [ 23 ], we counted the number of FASTA protein entries, HMM models, or website database size reports (for RAST and InterProScan), depending on the tool. Figure 2 a shows that MicrobeAnnotator and DRAM have orders of magnitude more entries than the other tools, and MicrobeAnnotator (~ 350 million) has almost three times the number of entries compared to DRAM (~ 121 million).…”

“…The methods used to search query protein sequences against these databases vary in complexity, comprehensiveness, speed, scalability, and results. For example, the National Center for Biotechnology Information (NCBI) Prokaryotic Genome Annotation Pipeline (PGAP, [ 20 ]), Prokka (prokaryotic annotation) [ 21 ], RAST (Rapid Annotations using Subsystem Technology) [ 22 ], and DRAM (Distilled and Refined Annotation of Metabolism) [ 23 ] start from genomes and predict genes and proteins, tRNAs, rRNAs, and perform functional annotation of the predicted proteins. Others such as InterProScan [ 12 ] and EggNOG-Mapper (the evolutionary genealogy of genes: Non-supervised Orthologous Groups) [ 11 ] start from already predicted protein sequences and perform functional annotations using mostly Blast (Basic Local Alignment Search Tool) [ 24 ], Diamond (double index alignment of next-generation sequencing data) [ 25 ], or HMMER [ 26 ] as search tools.…”

Minnesota peat viromes reveal terrestrial and aquatic niche partitioning for local and global viral populations

“…VIBRANT [50] and DRAM-v [51] were used to identify putative AMGs in the vOTU sequences. VIBRANT was run (using standard settings) on all SPRUCE viral contigs identified by either VirSorter or DeepVirFinder, resulting in 2,802 vOTUs that were used for this analysis.…”

“…Although we now have an array of laboratory and bioinformatics methods for soil viral ecology [7,15,23,31,34,[46][47][48][49][50][51], we lack a thorough comparative understanding of these approaches and best practices. As one specific example, viral size-fraction metagenomes (viromes) from grassland and agricultural soils have been shown to be substantially enriched in viral and ultrasmall cellular organismal DNA, compared to total metagenomes that tend to be more enriched in DNA from cellular organisms too large to easily pass through the 0.2 µm filters used for viral enrichment [35,52].…”

Metagenomics and quantitative stable isotope probing offer insights into metabolism of polycyclic aromatic hydrocarbons degraders in chronically polluted seawater

“…Iron-related transport and storage systems were identified using FeGenie(53). Additional functional analysis was also done with METABOLIC version 1.3(54) and DRAM(55).Read recruitment from different samples to the MAGs and viral contigs was analyzed with Anvi'o (39) using the Q2Q3 setting. This setting ignores the 25% lowest covered and 25% highest covered positions within the MAG when calculating mean coverage to avoid bias due to islands or highly conserved genes.…”