Editorial: Advances in optical methods for biological sample characterization

Optical methods have quietly become one of the most productive points of contact between physics and biology. The proper use of laser–tissue interactions can reveal the orientation of a single collagen fibril, the redox state of a mitochondrion, or the vibrational fingerprint of a virus, often without staining, fixing, or otherwise disturbing the sample under study. The four contributions gathered in this Research Topic, Advances in Optical Methods for Biological Sample Characterization, sample that breadth directly: a mechanistic study of second harmonic generation in the cornea, a critical review of the trade-offs built into fluorescent labelling, a forward-looking perspective on the field as a whole, and a plasmonic biosensing platform built to tell two closely related viruses apart. Read together, they show how mature certain optical techniques have become, and how much interpretive and instrumental work still remains.

We opened this Research Topic with a deliberately wide scope, inviting work on bioimaging, biosensors, optical tweezers, particle tracking, and AI-assisted data analysis. In our own Perspective (Avila et al.), we used these five domains as a scaffold for a broader argument: that the next substantial gains in optical biological characterization are likely to come less from any single modality reaching a new performance ceiling, and more from combining modalities, from tighter integration with computation, and from a sharper distinction between what an optical signal shows and what it is generally assumed to mean. The three remaining contributions to this Research Topic each speak to that argument from a different angle.

Boutkhil et al. take on a problem that has limited the clinical translation of second harmonic generation (SHG) microscopy for some time: backward-detected SHG signal, the only geometry realistically available for in vivo imaging, has historically been harder to interpret than the cleaner forward signal obtained from thin, ex vivo sections. Imaging intact porcine corneas in a custom pressure-controlled chamber, the authors show that the periodic high- and low-intensity bands long read in backward images as discrete “collagen bundles” are in fact an artifact of collagen crimp, the natural waviness of corneal lamellae. Raise the intraocular pressure and the fibres straighten; the bands fade in step. The authors are candid that a complete quantitative model linking the forward-to-backward signal ratio to tissue structure remains out of reach, but the structural mechanism they identify should change how the next-generation of corneal SHG images, and the diagnostic claims built on them, are read.

A second axis of the same general problem, quite literally whether to label a sample at all, runs through the review by Cuevas-Galindo and Jiménez-Sánchez. Small-molecule fluorescent probes are indispensable tools, but the authors build a sustained case that the same properties that make a probe bright and selective, high-affinity binding, lipophilic partitioning, excited-state reactivity, also make it an active participant in the biology it is meant to report on. Their discussion of calcium indicators, whose buffering capacity measurably reshapes the signalling kinetics they are used to track, gives a concrete anchor to an argument that can otherwise stay abstract. The review does not call for abandoning labels; it positions label-free approaches, autofluorescence imaging of NAD(P)H and flavins, fluorescence lifetime imaging, as a benchmark against which any probe’s effect on the underlying physiology should be checked, and it surveys middle-ground strategies, including HaloTag and SNAP-tag constructs, bioorthogonal fluorogenic chemistry, and ratiometric sensing, that try to narrow that gap.

The fourth contribution moves from imaging to sensing. Guilcapi et al. describe a plasmonic metasurface, patterned with Y-shaped nanocavity arrays and tuned to 785 nm excitation, that supports surface-enhanced Raman spectroscopy (SERS) sensitive enough to discriminate human from murine norovirus particles by their vibrational signature alone. Owing to the structural similarities between the two viruses, this result has implications beyond a simple proof of concept: principal component analysis of the SERS spectra recovers two components accounting for over 98% of the total variance, enough to separate the species reliably with no antibodies, no PCR, and no labelling step of any kind. Given how disruptive norovirus outbreaks remain for food safety and clinical care, a rapid, label-free screening route of this kind is an appealing complement to existing diagnostics, and the underlying nanofabrication approach should generalize to other closely related pathogens.

Taken together, these four contributions converge on a theme that recurs across biophotonics more broadly: an optical signal is rarely as simple to interpret as it first appears, whether because the geometry of signal Research Topic changes what a “bundle” or a “band” actually represents, because the probe used to generate contrast measurably perturbs the system it is meant to probe, or because telling two similar biological targets apart requires statistical tools as much as photons. None of these four pieces resolves those tensions outright, but each narrows one of them, by correcting a long-standing source of misreading, by naming a bias that is easy to overlook, by showing that closely related biological samples can be distinguished without labels, and by setting out a framework against which the field’s broader progress can be measured. With this Research Topic now closed, we thank the authors and reviewers who contributed to it, and we hope it serves as a useful reference point for further work at the interface of optics and biology.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. Claude (Sonnet 4.6) was used in preparing this manuscript—particularly, for drafting assistance that was then substantially revised.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited and reviewed by: Marco Peccianti, Loughborough University, United Kingdom