Biological research data are being generated at an unprecedented rate. Although visualization and modeling are critical to understanding biological research data, the size and complexity of these data pose challenges to effective exploration and interpretation. Existing solutions often remove potentially vital information or necessitate expertise in programming methods and data-processing techniques, which slows down discovery. Additionally, researchers commonly generate multiple omics data types for the same study. Integration of these complex, disparate datasets requires access to and understanding of databases of known biological pathways, knowledge of robust data-preprocessing techniques and statistical methods, and programming capability. Gaps in any one of these capabilities can lead to irreproducible results or heightened time requirements.
The Environmental Molecular Sciences Laboratory developed a framework in the form of a web application called the Multiomics Analysis Portal (MAP) where users can upload nuclear magnetic resonance (NMR) metabolomics, gas chromatography-mass spectrometry (MS) metabolomics and liquid chromatography-MS metabolomics and lipidomics, liquid chromatography-MS proteomics (labeled or label-free), and RNA-sequence (RNA-seq) transcriptomics data for exploration. This workflow starts with datasets consisting of quantified biomolecule relative peak intensities or spectral counts (RNA-seq). The capabilities in the workflow are implemented as modules with available and appropriate metrics, normalizations, and methods determined on the backend of the web application based on data characteristics rather than requiring expertise in biostatistics. At the end of the workflow, a report of analysis steps, figures, data, and code for reproduction are available for download.
Supporting the Data Transformations Integrated Research Platform (IRP), this tool handles a variety of data types that are generated from experiments in other IRPs. MAP takes in data after quantification of biomolecule relative peak intensities or spectral counts.
This includes data generated from:
- Labeled and unlabeled proteomics
- MS-based lipidomics and metabolomics
- NMR-based metabolomics
- RNA-seq data
- FT-MS (Fourier transform mass spectrometry) data (coming soon with the release of FREDA)
MAP tools include:
- PMart for analysis of a single dataset (excluding FT-MS data)
- iPMart for analysis and integration of multiple datasets generated from the same samples (excluding FT-MS)
- MODE for interactive visualization of data, including statistical results from PMart or iPMart sessions
- FREDA for exploratory data analysis of FT-MS data (coming soon)
- SLOPE for guided machine learning (anticipated release in Spring 2024)
The Multiomics Analysis Portal web application is available online. The application is hosted on the cloud, so requires a NEXUS login to access it.
Tips for Success
Check out the supporting resources:
- Documentation available via the MAP web application
- EMSL LEARN webinar on MAP capabilities
- Paper describing the underlying R code functionality for the PMart app, example graphs
These are the data types the current apps in MAP support:
- NMR metabolomics
- GC-MS metabolomics
- LC-MS metabolomics
- LC-MS lipidomics
- LC-MS/MS proteomics (labeled or label-free)
- RNA-seq transcriptomics