Most proteomic endeavors have focused on identifying and quantifying expression levels of low abundance proteins. These queries are necessary but not sufficient, as the assessment of proteins is incomplete without simultaneously shining light on post-translational modifications (PTMs) and how they augment the proteome. PTMs decorate translated proteins in a dynamic fashion, and the multitude of possible forms constitute the “epi-proteome,” in much the same way as epigenetic marks and DNA define the epigenome.
The most information-dense PTM is glycosylation, where various glycans (oligosaccharides) are attached to amino acid residues in a translated protein. One of four major classes of biological macromolecules found in all lifeforms, glycans play a major role in regulating protein folding, signaling, and numerous cellular functions. Aberrant glycosylation is increasingly being recognized in conditions such as cancer, inflammation, autoimmunity and neurodegenerative diseases. The science of how and why glycosylation introduces microheterogeneity within individual proteins has been on the fringe during the genomic era, but is poised to refine how we understand, intervene and treat many diseases. A major hurdle to translating this knowledge to the clinic lies in the hard-to-scale workflows and data interpretation of mass spectrometry, the only tool precise enough to measure both the composition and location of PTMs at a large scale.
This is where InterVenn comes in. InterVenn’s non-traditional approach to epi-proteomics is a marriage of mass spectrometry and AI. Powered by large, human-curated mass spectrometry datasets, we have developed advanced machine learning and artificial intelligence algorithms in the quest to industrialize mass spectrometry. The lack of such software has historically kept mass spectrometry an academic and small-scale pursuit. In contrast, InterVenn’s technology scales mass spectrometry to the platform level, opening new vistas in the depth and breadth at which proteomic and epi-proteomic data can be assessed and applied to the benefit of human health.
The scientific literature is rife with individual connections between protein glycosylation and disease, and InterVenn’s overarching goal is to bridge these connections with new discoveries to build highly accurate diagnostics and companion diagnostics, uncover scores of prognostic and predictive biomarkers, and characterize novel therapeutic targets. We believe that the stochastic yet decipherable nature of PTMs will complement the template-driven genomic approach to enable truly personalized medicine of the 21st century.