Help

This is Shiny GATOM: a web-service for omics-based identification of regulated metabolic modules in atom transition networks. It is an updade of Shiny GAM web-service (http://www.ncbi.nlm.nih.gov/pmc/articles/pmc4987878/) web-service.

Overall the analysis consists of two major steps:

• Step 1: Creating a metabolic network specific to the provided omics data.
• Step 2: Finding a connected subnetwork that contains the most significant changes.

Reset all button allows to start the analysis from the beginning, clearing the input data sets.

Example data

Three example datasets are embdeded in the web-service.

Example DE for genes and Example DE for metabolites checkboxes load transcriptional and metabolomic data respectively for M0 vs M1 macrophage activation comparison Jha et al, Immunity 2015. The raw files can be downloaded from here and here.

“Example DE for lipids” checkbox loads the example lipidomics data for WT mice control vs high fat diet comparison from ST001289 dataset. The raw file can be downloaded here.

We also provide example data with differential gene expression for human MCF10A cell line treated with 2DG (from GSE59228).

Network construction

First, slect an organism with which you will be working from the Select an organism dropdown menu. Only reactions possible in the selected organism are used. Currently supported organisms are Homo sapiens, Mus musculus, Arabidopsis thaliana and Saccharomyces cerevisiae.

The next step is to upload differential expression (DE) data for genes and/or metabolites. GATOM can be run using either gene DE data or metabolite DE data or both data types. Each DE dataset must be in a separate text file (comma-, tab- and space- separated files, archived files and Microsoft Excel Open XML Spreadsheet files are supported too). The first line of each file must contain a header with the column names. The following columns are expected:

• “ID”: RefSeq mRNA transcript ID, Entrez ID or symbol for genes; HMDB ID, KEGG ID or ChEBI ID for metabolites; SwissLipids ID, LipidMaps ID, ChEBI ID or Species name for lipids (supports LipidMaps and SwissLipids nomenclature). Multi-annotated data with IDs separated with “/// ” is also supported
• “pval”: Differential expression p-value (non-adjusted).
• “log2FC”: Base 2 logarithm of the fold-change.
• “baseMean” (for genes): average expression level.
• “signal” (optional): ID of the measured entity, such as probe ID for gene expression microarrays and ion ID for metabolite data, if absent, signal column will be generated based on unique “pval” and “log2FC” combinations.
• “signalRank” (for genes, optional): ranking of the genes by average expression, if absent, the values are generated based on “baseMean” column.

Any other columns will be copied to a network as node or edge attributes.

After files are uploaded, a file summary is displayed. In the bottom of the summary rows with IDs that were not mapped to the IDs used in network (e.g. Entrez for mouse) are shown. Verify that the files were parsed correctly.

Next, a network can be constructed. There are thee major network types: 1) based on KEGG database, 2) based on Rhea database, 3) subset of Rhea reactions involving lipids. Additionally, two network topologies are supported: atom- and metabolite-level. In atom-level topology, network nodes correspond to individual carbon atoms of compounds, and an edge connects to atoms if there is a reaction with such atom transition. In metabolite-level topology nodes correspond to metabolites, and two metabolites are connected if there is a reaction where at least on carbon atom is transitioned between the two metabolites.

Click Step 1: Make network to finish this step.

Scoring network

After making a network, the statistics on network structure will be shown.

On the scoring panel you can find the amount of positive gene and/or metabolite signals as well as corresponding alpha coefficient of the BU.

The size of the module can be controlled by changing scoring parameters. Increasing them will relax scoring thresholds and will result in a bigger module. User can choose one of the following scoring parameters – “Number of positive metabolites/genes”, “P-value threshold” or “FDR” or they can avoid using either metabolites or genes for scoring altogether with clicking “Don't use metabolites/genes for scoring”. When adjusting one parameter, others will change simultaneously.

Finding a module

After scoring of the network, you can find a connected module that contains the most significantly changed genes and metabolites. Internally, this is done by first scoring nodes and edges based on their p-values in such way that positive scores correspond to significant p-values and negative scores correspond to insignificant changes. The problem of finding a connected subgraph with the maximum weight is solved with Virgo solver (https://github.com/ctlab/virgo-solver).

Click Step 2: Find module button to find a module in the network. The module will be shown on the right panel.

Graph legend

We use the following scheme:

• Red nodes and edges are up-regulated (log2FC > 0).
• Green nodes and edges are down-regulated (log2FC < 0).
• Blue nodes and edges don't have log2FC values.
• Bigger size of nodes and width of edges means lower p-values.

Saving the module

You can download the module in several formats by clicking corresponding links:

• SVG file contains the vector image of the module, as it appears on the web-site,
• XGMML file contains module that can be imported into Cytoscape (see Importing to Cytoscape section below).
• XLSX file contains simple table of metabolites and reactions in the module, along with corresponding fields (logPval, log2FC, …).

Postprocessing

Next, reactions without highly changing genes but with high average expression can be added. Sometimes, in case of atom-level topologynthe same metabolite can appear multiple times in the module via different atoms. In such cases it is useful to either connect atoms belonging to the same metabolite with edges or to collapse them into one vertex. That can be done by choosing one of the action in “Actions with atoms”.

Functional annotation of obtained modules by KEGG and Reactome metabolic pathways will be calculated automatically and pathways can be highlighted on the module by choosing the name of the pathway in “Select a pathway to hightlight”. For detailed information of the pathway enrichment user can click on the “Pathway enrichment details”.

Interpreting the results

The result of the analysis is a set of connected most regulated reactions. These reactions are associated with strongly differentially changing enzymes and metabolites or are closely connected to such reactions which implies regulated flux. Inside the module one would typically consider small groups of particularly significant reactions or individual reactions close to the center of the module. Such groups or individual reactions are implied to be of biological importance for the considered process. These can lead to potential experimental designs including labeling, media perturbation or gene knockout/knockdown experiments. For examples see Network integration of parallel metabolic and transcriptional data reveals metabolic modules that regulate macrophage polarization and Mitochondrial Phosphoenolpyruvate Carboxykinase Regulates Metabolic Adaptation and Enables Glucose-Independent Tumor Growth papers.

Importing module into Cytoscape

XGMML file with module can be imported into Cytoscape for consecutive exploring and processing. For the best experience we recommend to use GAM's VizMap style for Cytoscape that can be downloaded here.

Running the analysis in R

GATOM analysis can be done manually in R. For this check out gatom R package.