Global stoichiometry conservation architecture of gene expression profiles revealed by cellular Raman spectra

Cells globally change their molecular profiles to adapt to their biological contexts and external conditions. Despite the complexity of molecular compositions, the changes in molecular abundance are often correlated. However, biological principles behind the low dimensionality of global molecular profiles in cells remain to be characterized.

Vibrational spectroscopy might help investigate the global constraints because it provides the information reflecting comprehensive molecular composition of target cells. Recently we have shown that cellular Raman spectra are quantitatively linked to multiple types of omics. The correspondence allowed us to infer global changes in omics from Raman spectra. Importantly, the Raman-based omics estimation was performed using dimension-reduced Raman spectra.

In this study, examining the correspondence between proteomes and Raman spectra of Escherichia coli, we characterize core and peripheral stoichiometrically-conserved groups in the proteomes. With methods in graph theory, we reveal a low-dimensional proteome structure constructed from global stoichiometry conservation relations of genes. This structure is reflected in major changes in Raman spectra. Furthermore, the centrality of the stoichiometry conservation architecture correlates with the essentiality and evolutionary conservation of genes. These results provide an omics-level interpretation of Raman spectra and uncover an architectural principle of global omics profiles.