Fes0500

All absorption data was taken on a Varian Cary 500 spectrometer. Rhodamine 6G emission measurements were taken on a Horiba Scientific Fluorolog spectrofluorimeter using a monochromated Xe lamp as the excitation source. CEP blend film and Stilbene-420 emission spectra were taken on a LaserStrobe spectrometer from Photon Technology International using a GL-3300 nitrogen laser and GL-302 dye laser attachment, also from Photon Technology International. Upconverted emission spectra were measured with the emission filtered by a 500 nm short-pass filter from Thorlabs, model FES0500, to prevent reflected excitation light from interfering with the measured emission signal. Laser power was measured with a 919P-003-10 thermopile sensor from Newport. Time-Resolved Single Photon Counting Data (in Extended Data section) was taken using excitation light generated by a Fianium SC400 supercontinuum fibre laser with wavelength selected by a Fianium AOTF system. Detection was measured via a photomultiplier tube connected to a Becker-Hickl SPC-130 system. All data was collected with signal count rate at <2% of excitation rep rate to ensure proper TCSPC statistics. All spectra were corrected for the spectral responsivities of the systems used for data collection.

A 488 nm diode laser (200 mW, Cobolt 06-MLD) was used for fluorescence excitation. The excitation beam was expanded through a 4f system (f₁ = 50 mm, LA1131-A-ML and f₂ = 300 mm, LA1484-A-ML, Thorlabs) and then coupled into the light path of probe laser. The fluorescence emission went through the same objective lens and was collected with a filter set (Excitation filter: FES0500, Thorlabs; Dichroic beam splitter: Di03-R405/488/532/635-t1-25×36, Sermock; Emission filter: FF01-525/30-25, Sermock). A CMOS camera (FLIR, Grasshopper3GS3-U3-51S5M) was used to capture the fluorescence images and the exposure time was set to 5 s for optimized contrast. Virus samples on silicon substrate were first imaged by fluorescence to confirm the single virus and then imaged with WIDE-MIP at the same position.

The comprehensive experimental setup of the proposed tomography system is demonstrated in Supplementary Fig. 2. This system employs an MCF (FIGH-350S, Fujikura) to dynamically modulate the output light field, controlled by an SLM (PLUTO-2, Holoeye), which further enables optical control of cell rotation. The core of the setup, the MCF-OCR, comprises the MCF and an opposing SMF (SM600, Thorlabs). The SMF has dual functions: it accommodates the reference beam used for calibrating the MCF, and it is also used as the waveguide for the near-infrared fiber laser (Eylsa 780; Quantel), which is responsible for optical trapping. The output laser power from each optical fiber is kept under 40 mW to minimize the risk of photodamage and photothermal effects. A custom-built brightfield microscope is used to image the cell rotation process. We use a blue light-emitting diode (LED) (M455L4, Thorlabs) as the light source for bright-field imaging. The area of optical manipulation is magnified by a microscope objective (50×, 0.42 NA, Mitutoyo) and tube lens (TTL200, Thorlabs), and this magnified image is projected onto a recording camera (Ueye CP, IDS) via lens systems. To achieve clear and high-contrast imaging, short-pass filters (FES0500, Thorlabs) are positioned before the camera to eliminate scattered light from the laser beams used for optical manipulation.