Integrative meta-omics in Galaxy and beyond

Abstract

Background: 'Omics methods have empowered scientists to tackle the complexity of microbial communities on a scale not attainable before. Individually, omics analyses can provide great insight; while combined as "meta-omics", they enhance the understanding of which organisms occupy specific metabolic niches, how they interact, and how they utilize environmental nutrients. Here we present three integrative meta-omics workflows, developed in Galaxy, for enhanced analysis and integration of metagenomics, metatranscriptomics, and metaproteomics, combined with our newly developed web-application, ViMO (Visualizer for Meta-Omics) to analyse metabolisms in complex microbial communities.

Results: In this study, we applied the workflows on a highly efficient cellulose-degrading minimal consortium enriched from a biogas reactor to analyse the key roles of uncultured microorganisms in complex biomass degradation processes. Metagenomic analysis recovered metagenome-assembled genomes (MAGs) for several constituent populations including Hungateiclostridium thermocellum, Thermoclostridium stercorarium and multiple heterogenic strains affiliated to Coprothermobacter proteolyticus. The metagenomics workflow was developed as two modules, one standard, and one optimized for improving the MAG quality in complex samples by implementing a combination of single- and co-assembly, and dereplication after binning. The exploration of the active pathways within the recovered MAGs can be visualized in ViMO, which also provides an overview of the MAG taxonomy and quality (contamination and completeness), and information about carbohydrate-active enzymes (CAZymes), as well as KEGG annotations and pathways, with counts and abundances at both mRNA and protein level. To achieve this, the metatranscriptomic reads and metaproteomic mass-spectrometry spectra are mapped onto predicted genes from the metagenome to analyse the functional potential of MAGs, as well as the actual expressed proteins and functions of the microbiome, all visualized in ViMO.

Conclusion: Our three workflows for integrative meta-omics in combination with ViMO presents a progression in the analysis of 'omics data, particularly within Galaxy, but also beyond. The optimized metagenomics workflow allows for detailed reconstruction of microbial community consisting of MAGs with high quality, and thus improves analyses of the metabolism of the microbiome, using the metatranscriptomics and metaproteomics workflows.

Keywords: Bioinformatics; Galaxy; Integrated meta-omics; Metagenomics; Metaproteomics; Metatrascriptomics.

Fig. 1

Workflows for meta-omics. The integrated analysis of meta-omics contains a MetaG, MetaT and MetaP workflow. MetaG includes data preprocessing steps with quality control and trimming, followed by assembling, binning and taxonomically annotation of the MAGs. Open reading frames (ORFs) and nucleotide sequences are predicted by FragGeneScan. Functional annotation is performed by InterProScan and dbCAN-HMMER. The predicted ORFs and nucleotide sequences are further used in the MetaP and MetaT workflow; hence, the MetaG serves as the base analysis and the MetaT and MetaP are mapped onto the MetaG. After preprocessing the data and rRNA removal, the predicted nucleotide sequences from the MetaG workflow are used for the mRNA quantification and mapping by Kallisto, as well as for MaxQuant in the MetaP workflow

Fig. 2

ViMO visualizations. A ViMO produces bar plots to visualize the gene counts and abundances of KEGG-pathways in the different bins, here filtered to pathways in energy metabolism. For metagenomics, all timepoints are used, while for metatranscriptomics and metaproteomics, only the first timepoint is shown here and the user can select which sample/timepoint to visualize. In addition, ViMO displays heatmaps with all timepoints within one graph for metatranscriptomics and metaproteomics to visualize temporal changes (data not shown). B ViMO calculates the module completion fraction (mcf) for all KEGG modules (x-axis; only a subset displayed here) and MAGs (y-axis) and thus visualize the metabolic potential of each MAG. The set of visible modules can be filtered to selected KEGG pathways for in-depth exploration

Fig. 3

Annotated KEGG-maps. In ViMO, when KEGG-pathways are selected (top, filtered to pathways in carbohydrate metabolism), a KEGG-map is downloaded and annotated with abundances of expressed genes for the selected MAG. Here is shown the Glycolysis/Gluconeogenesis pathway of MAG001, a bacterium from the Tissierellia class in the SEM1b community, annotated with metaproteomic abundances ranging from low-abundant (0 LFQ; light yellow) to high-abundant (4e9 LFQ; dark red); blue enzymes are not detected in the metaproteome for this MAG

Fig. 4

Optimized metagenomic workflow. We have created an optimized MetaG workflow to improve the quality of the MAGs. This is achieved by assembly and binning of the reads individually, in parallel to a co-assembly, and combined and dereplicated to exclude redundant MAGs before bin annotation, gene prediction and functional annotation. Two samples S1 and S2 are shown as an example. Differences to the original MetaG workflow are highlighted in yellow