The significance of controlled single-cell rotation extends well beyond a clever laboratory demonstration. When a cell can be turned at will, researchers can perform true three-dimensional imaging with conventional microscopes, capturing morphology and internal architecture from multiple viewing angles. Rotation also matters for mechanobiology, the study of how cells sense and respond to mechanical forces, because orientation relative to a substrate or a neighboring cell influences how forces are transmitted and interpreted. Controlled rotation can further support investigations of cell-cell interactions, intracellular dynamics, and developmental processes in which the spatial arrangement of cellular components is central to function.

Yet programmable rotation has long been an elusive goal. Optical tweezers, magnetic manipulation, electric fields, and acoustic techniques have each made substantial contributions to single-cell control, but most existing approaches depend on specially engineered beam structures, microfabricated mechanical components, or intrinsic asymmetries in the objects being manipulated. Spherical particles and many living cells present a particular difficulty: because they are geometrically symmetric, there is no natural feature that defines a rotation axis. Dynamically switching between different rotation modes within a single platform has remained largely unexplored, which has limited the flexibility of previous systems and confined many of them to narrow, pre-set operating regimes.

The new platform is built on a plasmonic substrate composed of gold nano-islands functionalized with bovine serum albumin, or BSA. When laser light illuminates these nanostructures, they generate localized temperature gradients that drive strong thermo-osmotic flows along the solid-liquid interface. Thermo-osmotic flow is a well-characterized phenomenon in which a temperature gradient at a boundary produces directed fluid motion, and plasmonic substrates are especially effective at creating steep, highly localized gradients because the nanostructures concentrate optical energy into nanoscale volumes. The BSA functionalization plays a complementary role in stabilizing cells at the interface, while polyethylene glycol molecules in the surrounding solution generate depletion forces that gently confine cells near the substrate surface, precisely where the thermo-osmotic flow is strongest.

Control over the flow field comes from a spatial light modulator, a device that can dynamically project programmable laser patterns onto the substrate. By reshaping these patterns in real time, the researchers tailor both the temperature distribution and the resulting flow field, and with them the forces and torques acting on individual cells. This is the conceptual heart of the platform: the rotation axis is not dictated by the geometry of the object but is instead written into the engineered flow field itself. The same cell, sitting in the same spot, can be made to spin in entirely different ways depending only on how the laser light is patterned.

The team first validated the concept using spherical polydimethylsiloxane particles, an ideal test case precisely because a sphere possesses complete geometric symmetry and its rotation axes cannot be defined by shape. Numerical simulations revealed that a single Gaussian laser spot positioned asymmetrically relative to a particle creates a highly non-uniform thermo-osmotic flow field around it. That asymmetry in the flow produces a hydrodynamic torque of sufficient magnitude to drive stable rotation. Crucially, by simply moving the position of the laser spot, the direction of the torque can be tuned continuously, allowing the particle to rotate about different axes. Experimental observations confirmed the theoretical predictions and demonstrated arbitrary-axis rotation of spherical particles, establishing that the axis of rotation is defined entirely by the engineered flow field rather than by particle geometry.

With the principle demonstrated on synthetic spheres, the researchers extended the approach to biological cells, using yeast as a model system. They showed that two distinct rotational modes can be selectively activated through optical pattern engineering. Under a single Gaussian laser spot, the induced thermo-osmotic torque drives the cell to rotate around its major axis, the long axis of the cell body. When the laser pattern is reconfigured into a half-ring shape, an optical torque joins the thermo-osmotic torque, and the combined action reorients the cell and drives continuous rotation around its minor axis. The two modes are therefore not separate operating regimes requiring different hardware; they are two settings of the same optical interface.

Perhaps the most striking capability of the platform is the speed and cleanliness of the switching between these modes. In the reported experiments, the team maintained stable major-axis rotation with a Gaussian beam, then changed the illumination pattern to a half-ring configuration. The cell transitioned into minor-axis rotation in less than one second. The entire process occurred without mechanical movement, sample repositioning, or physical contact, relying solely on optical reconfiguration. For live-cell imaging, this matters enormously: cells are fragile, and any approach that avoids contact, labeling, or chemical modification preserves the native state of the specimen while it is being manipulated.

To understand why the system works so well, the researchers combined multiphysics simulations with electromagnetic calculations. Their analysis showed that major-axis rotation is predominantly driven by thermo-osmotic torque, whereas minor-axis rotation arises from the synergistic coupling of thermo-osmotic and optical torques. Together, these torques create a continuous torque landscape that sustains rotation throughout the entire cycle without trapping the cell in stable equilibrium states. In other words, the optical patterns are designed so that at no point in the rotational cycle does the cell find a configuration where the torques vanish and motion stops; the driving torque persists around the full revolution, which is what enables continuous, stable spinning.

The practical advantages of the platform are considerable. It operates at relatively low optical power, which reduces the risk of photothermal damage to living specimens. It requires no labeling or modification of cells, making it a genuinely label-free technique. It offers exceptional programmability through dynamic optical pattern generation, and unlike many conventional rotational manipulation techniques, it does not depend on complex beam shaping or specialized particle geometries. The researchers believe the technique could become a powerful tool across a broad range of biological and biomedical applications. By integrating with confocal microscopy and other advanced imaging modalities, the platform may enable label-free three-dimensional cellular imaging and reconstruction. It could also facilitate studies of cell mechanics, cellular interactions, developmental biology, and drug screening, where observing a cell from multiple orientations under controlled conditions would add a valuable new dimension to existing assays.

Looking forward, the team envisions combining opto-thermo-osmotic manipulation with emerging metasurface and integrated photonic technologies. Metasurfaces, which are engineered planar optical structures capable of shaping light with subwavelength precision, could replace bulky spatial light modulators with compact, chip-scale pattern generators. Such integration could lead to miniature platforms capable of high-throughput cellular manipulation and analysis, opening new opportunities for next-generation bioanalytical systems in which individual cells are imaged, rotated, and characterized automatically. If that vision is realized, the ability to spin a single cell along any axis with nothing more than a reconfigured beam of light may become as routine a capability in the biology laboratory as focusing the objective lens is today.

Subject of Research: Programmable opto-thermo-osmotic rotation of single cells along arbitrary axes

Article Title: Programmable rotation of single cells along arbitrary axes via opto-thermo-osmotic torque

Article References: Programmable rotation of single cells along arbitrary axes via opto-thermo-osmotic torque. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: cell rotation, opto-thermo-osmotic torque, plasmonic substrate, gold nano-islands, thermo-osmotic flow, spatial light modulator, yeast cells, three-dimensional imaging, mechanobiology, optical manipulation, label-free imaging, Light Science & Applications

At the heart of the controversy is a technique the authors call SLRI, or spectral label-free Raman imaging, which combines spontaneous confocal hyperspectral Raman microscopy with a deterministic mathematical fitting procedure known as non-negative least squares, or NNLS. The method scans tissue point by point with a single 532-nanometer excitation beam, acquiring a full Raman spectrum at every pixel across thousands of spatial positions and multiple depth planes. Each measured spectrum is then fitted against a library of 42 purified reference compounds, ranging from saturated and unsaturated fatty acids to cytochrome c and other biomolecules, producing spatial maps that show where each reference component contributes most strongly to the measured signal.

The critical letter, according to the authors, misread the scope of these claims. The reply is unambiguous on this point: the published article never asserted that the NNLS-derived maps represent absolute molecular concentrations, nor that all 42 analytes had been definitively and independently identified. Instead, the paper explicitly describes the maps as reference-constrained relative distributions, and the small-molecule images as Raman-inferred, model-constrained representations. The authors lay out a clear analytical hierarchy to make the distinction concrete: a measured Raman spectrum feeds into a deterministic, reference-constrained NNLS coefficient, which in turn generates a spatial coefficient map. No step in the Drosophila study converts those coefficients into milligrams per milliliter or moles per liter. A separate concentration-calibrated implementation exists in the team’s earlier qRamanomics work, published in Cell Reports Methods in 2023, but that calibration was claimed only for selected major biomolecular classes, not for all 42 references.

Where the published text used shorthand such as a component being enriched in a region, the authors explain, the intended meaning is a relatively higher distribution of the corresponding reference-constrained spectral contribution, not a calibrated concentration increase. They also address a related criticism about statistics: violin plots in the supplementary figures quantified only three representative components, unsaturated fatty acids, saturated fatty acids, and oxidized cytochrome c, at the whole-testis level. These were chosen as illustrative examples of signal trends, the authors say, not because the imaging was limited to three molecules. Crucially, they acknowledge that pixels within a single specimen are spatially correlated measurements and were never treated as independent biological replicates, so pixel-level statistical tests were not used as evidence of biological replication.

On the question of whether the computational workflow was reconstructible from the public record, the authors concede the Advanced Science article could have linked the code and study-specific data more explicitly, but they reject the charge that the pipeline was opaque. The same MATLAB qRamanomics processing and NNLS codebase used in the study was previously published and archived on Zenodo, and the principal new analytical input was the study-specific 42-reference library. To close the loop, the team has now released a set of files that make the data lineage explicit: the normalized 42 reference spectra, a representative raw hyperspectral Z-plane containing 30,771 measured spectra on a 263 by 117 grid with 1,114 Raman-shift channels per spectrum, a numerical NNLS coefficient vector for a cytochrome c map, and the corresponding rendered TIFF image. Reshaping the coefficient vector to a 600 by 686 array reproduces the published image with a Pearson correlation of approximately 0.9999, a striking demonstration that the colorful figures trace directly back to raw pixel-resolved spectral measurements rather than to stylized graphics.

Perhaps the most technically substantive portion of the reply concerns identifiability, the question of whether 42 overlapping spectra can be reliably separated at all. The authors report new diagnostics on the exact reference library after interpolation onto the measured Raman axis and application of the fitting window from 400 to 3100 inverse centimeters, excluding the spectroscopically silent region between 1800 and 2700. The 42-by-42 design matrix is full rank, meaning no reference spectrum is an exact linear combination of the others, with 500 fitting variables available for 42 components. Yet the team does not oversell this result. The 2-norm condition number of the unit-area-normalized matrix is 391.5, dropping to 242.6 after additional L2 column normalization, and pairwise correlation and hierarchical clustering analyses reveal substantial collinearity among several chemically related references. The authors explicitly state that these diagnostics are not proof that all 42 components are mutually orthogonal or unconditionally identifiable; rather, they quantify the model dependence that motivated the cautious Raman-inferred terminology in the original paper. They also dismiss a tempting shortcut: refitting a noise-free synthetic mixture with the same design matrix would be mathematically circular and therefore meaningless as validation.

The reply also clarifies sample preparation, which bears directly on whether long acquisitions could have been corrupted by biological motion. The Raman-imaged testes were fixed in 4 percent paraformaldehyde for 20 minutes, washed three times in phosphate-buffered saline, embedded in agarose, and maintained in PBS during acquisition. The specimens were therefore not living tissues undergoing cell migration, differentiation, or metabolite turnover during scans that can nominally run for many hours. The authors further explain that the punctate and ring-like patterns highlighted in the study were identified directly within individual XY planes, comparing measured spectra from peripheral punctate loci and central diffuse loci within the same structure, so variation in z-step size cannot generate the observed patterns. Three-dimensional renderings simply stack consecutively acquired planes recorded in the same microscope coordinate system; no intensity-based registration between separately acquired channels is involved, because every molecular coefficient at a given coordinate derives from the same measured spectrum.

On the modality question, the authors note there is no experimental ambiguity: the study describes spontaneous confocal Raman microscopy with single-beam 532-nanometer excitation, point scanning, full-spectrum acquisition, and EMCCD detection, features that distinguish it from stimulated Raman scattering. The critical letter itself, they observe, reached the same classification, so the objection reduced to a citation-placement issue rather than a claim that the wrong technique was used. The authors also take aim at the conceptual figure accompanying the letter, pointing out that its microscope illustration corresponds to no confocal Raman instrument, its spectral traces are schematic drawings rather than experimentally acquired spectra, and its laser beam is drawn blue when 532-nanometer light is green. If AI-assisted tools generated the figure, they add, appropriate disclosure under journal policies would be important.

The broader stakes of the exchange concern the role of artificial intelligence in spectroscopy. The letter proposed an AI-ready multimodal spatial-biology framework as an alternative, but the authors argue that a conceptual schematic is not evidence of analytical superiority. AI may prove valuable for Raman denoising, representation learning, and multimodal integration, they concede, but it is neither necessary nor evidentiary for validating a deterministic study whose reproducibility rests on fixed inputs, a specified reference matrix, documented preprocessing, and NNLS optimization, all of which yield identical numerical outputs given identical inputs and code. They challenge the correspondent to present one or two core experimental datasets to substantiate the proposed framework.

The authors close by situating the Raman maps within a wider web of orthogonal evidence: two independent Cyp4ae1 RNAi lines consistently produced the spermatogonial differentiation arrest phenotype, while LD540 and TOM20 staining, ATP and NADP+/NADPH assays, and non-targeted metabolomics independently documented lipid accumulation and mitochondrial abnormalities. Bulk assays do not prove microscopic localization of any individual Raman component, the authors acknowledge, but they support the broader metabolic remodeling within which the spatial Raman-inferred patterns are embedded. With the full datasets now posted publicly, including the raw hyperspectral layers, reference library, and diagnostics, the team argues that the study rests on real pixel-resolved data and a documented, deterministic pipeline, and that the appropriate standard for evaluating any spectral imaging claim is a minimum evidentiary framework in which the strength of the interpretation never exceeds the strength of the underlying spectra.

Subject of Research: Label-free Raman imaging methodology and evidentiary standards for spectral map interpretation in biomedical research

Article Title: From Spectral Maps to Biomedical Evidence: A Minimum Evidentiary Framework for Label‐Free Raman Imaging (Reply to Letter‐to‐the‐Editor)

Article References: Li, J., Zhao, X., & Yu, J. (2026). From Spectral Maps to Biomedical Evidence: A Minimum Evidentiary Framework for Label‐Free Raman Imaging (Reply to Letter‐to‐the‐Editor). Advanced Science, Article e77930. https://doi.org/10.1002/advs.77930

Image Credits: AI Generated

DOI: 10.1002/advs.77930

Keywords: Raman spectroscopy, NNLS, label-free imaging, Drosophila, spectral unmixing, reproducibility, qRamanomics, hyperspectral imaging, condition number, biomedical evidence, Cyp4ae1, spatial biology

Raman spectroscopy is built on a simple physical principle with powerful biological consequences. When laser light scatters off molecules, a small fraction of the photons exchanges tiny amounts of energy with molecular vibrations, and the resulting spectral shifts form a fingerprint of the chemical bonds present in the sample. Lipids, proteins, nucleic acids and carbohydrates all produce distinctive vibrational signatures, which means a Raman spectrum acquired from a living cell is, in effect, a quantitative chemical readout of its internal composition. Crucially, the method requires no dyes, no antibodies and no genetic modification, eliminating many of the artifacts and biases that complicate fluorescence-based approaches. The authors of the new work recognized that this chemical richness, combined with spatial resolution at the level of individual cells and subcellular compartments, could offer a window into senescence that no single molecular marker could provide.

The central innovation of RamanOmics lies in integration. Rather than using Raman imaging as a stand-alone imaging modality, the researchers systematically aligned vibrational spectra with gene-expression data obtained from matched tissue regions and single cells. Spatial transcriptomics maps messenger RNA abundances across intact tissue sections, preserving the geographic relationships between cells, while single-cell transcriptomics resolves expression states cell by cell. By overlaying these maps with label-free Raman images of adjacent or identical tissue sections, the team could ask a question that neither technology answers alone: which biochemical vibrational signatures correspond to which transcriptional states in the intact spatial context of aging or regenerating tissue?

The most striking finding to emerge from this analysis is a lipid-linked Raman signature associated with p21-positive cells. Cells expressing high levels of p21, a canonical enforcer of the senescence cell-cycle arrest, displayed a distinctive vibrational profile dominated by lipid features, suggesting that these cells undergo reproducible changes in their lipid composition or lipid storage as they enter senescence. This is consistent with a growing body of evidence that senescent cells accumulate neutral lipids and alter membrane composition, but the new work goes further by demonstrating that this lipid remodeling is spatially organized and detectable without any label. In principle, this means researchers could identify senescent cells in intact, unstained tissue by their chemical fingerprints alone, opening the door to quantitative mapping of senescence burden across organs, across ages, and across disease states.

Why would senescent cells show altered lipid signatures? Senescence is a metabolically demanding state. Cells in senescence secrete inflammatory cytokines, growth factors and matrix-remodeling enzymes, collectively known as the senescence-associated secretory phenotype, and they sustain this activity while remaining growth-arrested. Lipid droplets and membrane lipids are deeply involved in these processes, serving as energy reserves, signaling platforms and sources of secreted lipid mediators. A vibrational signature concentrated in lipid bands therefore plausibly reflects the reorganization of lipid metabolism that accompanies the senescence program. The alignment of this signature with p21 expression through transcriptomic overlay gives the spectral observation a molecular anchor, transforming what could be an ambiguous spectroscopic pattern into a biologically interpretable feature.

The relevance of the method extends beyond basic cell biology because senescence is double-edged. In aging tissues, the accumulation of senescent cells contributes to chronic inflammation, tissue dysfunction and a host of age-related pathologies, and pharmaceutical approaches called senolytics aim to remove these cells to extend healthspan. Yet in wound repair, senescent cells appear transiently and appear to be beneficial, promoting tissue regeneration, recruiting immune cells and stimulating progenitor activity before being eliminated. A tool that can distinguish and map senescence in both contexts, without labels that might bias the analysis, is therefore valuable for understanding when senescence is a driver of decline and when it is a constructive part of healing. The spatial dimension is essential here: the same cell type may be harmful in one tissue microenvironment and helpful in another, and only spatially resolved methods can capture that distinction.

The technical achievement of RamanOmics also highlights a broader trend in the life sciences, the convergence of physics-based imaging with computational genomics. Raman spectra are high-dimensional data, and extracting biological meaning from them requires machine learning and careful calibration against independent measurements. By treating spectra as molecular phenotypes and training the analysis on transcriptomic ground truth, the researchers effectively created a dictionary that translates vibrational features into biological states. This translational framework is what gives the approach its name and its potential generality. If the dictionary can be extended to other cell states, cell types and tissues, Raman imaging could evolve from a specialized physical technique into a routine omics platform, one that measures chemistry directly in intact tissue rather than inferring it from dissociated cells.

There are, of course, important questions ahead. Raman signals are inherently weaker than fluorescence, so imaging speed and sensitivity remain practical constraints, particularly for three-dimensional or large-field applications. Spectral assignments can be ambiguous, and the correspondence between vibrational features and specific lipid species or metabolic states will require continued validation against orthogonal chemical methods. Translating findings from experimental models of aging and repair to human tissue will demand demonstration that the lipid-linked signature of p21-positive cells is conserved across species and disease contexts. Nonetheless, the conceptual advance is clear. The study establishes that label-free vibrational imaging can carry biologically meaningful, transcriptomically anchored information about a complex and clinically significant cell state, and that this information can be placed precisely within tissue architecture.

The implications ripple outward to drug development and diagnostics. Senolytic and senomorphic therapies currently lack robust pharmacodynamic biomarkers, and clinical trials would benefit enormously from a way to measure senescent cell burden in tissue biopsies without relying on a patchwork of imperfect stains. A validated Raman-based readout of senescence could serve exactly this role, providing quantitative, spatially resolved, label-free assessment of whether an intervention is actually reducing the senescent population in a target tissue. Similarly, in regenerative medicine, where controlled induction of transient senescence may be a feature of successful healing, the technology could monitor whether engineered tissues or cell therapies recapitulate the beneficial phase of the senescence program without tipping into chronic, pathological accumulation.

What makes this work resonate beyond its immediate findings is the way it reframes an old biological question with new instrumentation. Senescence has been studied for decades through lens-based microscopy and molecular assays, each revealing a fragment of the phenomenon. RamanOmics stitches those fragments together by reading the chemistry of cells directly and aligning it with the genetic programs they express, all in the spatial context of aging and repair. If the framework proves broadly applicable, the vibrational landscape of tissue may become as routinely interrogated as its transcriptome, and the enigmatic cells that linger at the crossroads of aging and healing will finally be seen in full chemical detail, without ever having to touch them with a label.

Subject of Research: Label-free Raman imaging integrated with spatial and single-cell transcriptomics to identify molecular signatures of cellular senescence in aging and tissue repair.

Article Title: RamanOmics decodes the spatial vibrational–molecular architecture of senescence in aging and repair

Article References: Zhang, K., Chen, X., Monticolo, F., Sorrentino, S., Huang, H., Callahan, C., Qiao, Y., Zhou, J., Brodowska, S., Sapantzi, S., Qi, J., Wu, Y., Dang, T. N. S., Cao, Y., Kang, S., Viggiani, F., Ho, C.-K., Xu, Y., Kobayashi-Kirschvink, K. J., … Shu, J. (2026). RamanOmics decodes the spatial vibrational–molecular architecture of senescence in aging and repair. Nature Aging. https://doi.org/10.1038/s43587-026-01219-7

Image Credits: AI Generated

DOI: 10.1038/s43587-026-01219-7

Keywords: RamanOmics, cellular senescence, Raman spectroscopy, spatial transcriptomics, single-cell transcriptomics, p21, lipid signature, aging, tissue repair, senolytics, label-free imaging, Nature Aging