Unravelling Electronic Structure and Molecular Vibrations of Proteins in Virus Using Novel Correlated Plasmon‐Enhanced Raman Spectroscopy With Machine Learning

Why It Matters

CP‑ERS provides rapid, label‑free viral typing and deep electronic insight, potentially transforming diagnostics and virology research.

Key Takeaways

• CP-ERS uses gold quantum‑dot chip for enhanced Raman signals
• Detects quasielastic and inelastic electronic Raman scatterings in dengue proteins
• Machine‑learning model classifies dengue strains with 100% hold‑out accuracy
• Technique is non‑destructive, stable, and reproducible across samples
• Reveals glycosylation‑induced spectral changes in viral E‑proteins

Pulse Analysis

Raman spectroscopy has long been prized for its ability to map molecular vibrations, yet its inherently weak cross‑section has limited its use on viral proteins, especially the envelope (E) proteins that drive infection. The newly reported correlated plasmon‑enhanced Raman spectroscopy (CP‑ERS) overcomes this barrier by embedding highly oriented single‑crystalline gold quantum dots (HOSG‑QDs) onto a chip that supports low‑loss, tunable correlated plasmons. When a resonant laser excites these plasmons, the electromagnetic field is amplified and couples directly to the electronic and vibrational states of the virus, producing spectra that are both intense and reproducible.

Applying CP‑ERS to multiple dengue virus serotypes uncovered previously invisible quasielastic and inelastic electronic Raman scatterings, as well as distinct phonon modes linked to the E‑protein’s structural motifs. The technique also proved sensitive to post‑translational modifications; altered glycosylation patterns shifted specific peaks, offering a direct spectroscopic handle on viral maturation and antigenicity. These discoveries not only expand the fundamental understanding of viral protein physics but also provide a rapid, label‑free fingerprinting method that can differentiate serotypes without the need for antibodies or nucleic‑acid amplification.

The authors paired the rich CP‑ERS datasets with a supervised machine‑learning classifier, achieving 100 % accuracy on a hold‑out set and 93 % mean accuracy across five‑fold cross‑validation. This level of performance suggests that a portable CP‑ERS platform could serve as a point‑of‑care diagnostic, delivering results within minutes and reducing reliance on costly PCR infrastructure. Beyond dengue, the approach is adaptable to any pathogen or solid‑state material where electronic structure matters, positioning CP‑ERS as a versatile tool for biotech firms, pharmaceutical pipelines, and materials research labs seeking high‑throughput, non‑destructive analysis.