Proteogenomics studies involve multiple layers of omics data, from samples to features to sets of features such as functional enrichments. If not analyzed correctly, there is a risk of missing out on biological insights or overlook technical limitations. We developed nOmics with the intent of enabling rapid exploration across different data layers.
The software is developed in RShiny and leverages the Bioconductor ecosystem for its visualizations. It was used for exploration of a proteogenomics dataset in oat where the protein expression of two oat varieties infected by Fusarium graminearum was compared using transcriptome-based references to identify features distinguishing their infection response.
nOmics allows inspecting the data both globally using high-level interactive illustrations such as PCA, clustering and histograms and locally by zooming in on individual proteins to examine their expression patterns and sequence variations across references. Subsets of the data can easily be sliced out and used for calculating gene ontology enrichment and for identifying proteins of interest. nOmics was employed in the oat dataset where it was successfully demonstrated to reveal and handle technical anomalies and to identify both enriched gene ontology terms and proteins potentially linked to differing infection response across oat varieties.
The study demonstrates how nOmics allows for rapid inspection of multiple layers of proteogenomics data. It provides a nuanced overview of omics datasets, allowing for pinpointing of technical issues and rapid examination of functional relations.