OPTICAL MODULATION

Fluorescence, switched on command — background rejected by algorithm.

Scroll to descend

The specimen in the FFT domain, computed

GFP beta-barrel ribbon structure, PDB 1EMA

The switch · OMFP

Proteins that answer to light

Optically modulated fluorescent proteins drive their intensity up and down under a red-shifted co-illumination. Fluorescence becomes a controllable variable — not a constant.

The measurement · SAFIRe

Signal, recovered by algorithm

A fast Fourier transform of the time-series rejects every photon that does not modulate — which is to say, all of the background. Reference-free, background-free imaging: SAFIRe.

The engineering · SDM

New proteins, by design

Site-directed mutagenesis guided by computer-assisted 3D models of the chromophore environment. Structure first, sequence second.

Fig 1.1 — Specimen · GFP, transfected cells

Background-free imaging, via DSP

OMFPs — a novel class of fluorescent proteins uncovered by our team — hold the potential to revolutionize live-cell fluorescence imaging. Experimental data from diverse sources were systematically interconnected, and pivotal insights extracted from a sea of background noise.

Absolute reference-free and background-free cellular images were achieved by applying mathematical transformations — the fast Fourier transform — to time-series cellular fluorescence data, using bespoke algorithms in Python, Matlab, and Mathematica.

OMFPSAFIReFFT lock-inTime-series DSP

Engineering the proteins

Fluorescence intensity of OMFPs is precisely modulated via co-illumination with photons more red-shifted than the emission itself. New OMFPs are engineered through site-directed mutagenesis, drawing on intramolecular interactions between the chromophore and its environment via computer-assisted 3D modeling.

SDM3D modelingChromophore environment

Simulating the photophysics

Optical modulation and delayed fluorescence are modeled with multi-state rate matrices analogous to Jablonski diagrams — determining photophysical parameters previously too complex to assess. The simulation stack: NumPy and Pandas for computation, SciPy for optimization, Matplotlib for visualization, in an object-oriented paradigm.

Rate matricesJablonski-analogousPython · SciPy

GFP ribbon structure, PDB 1EMA — **Anchor publication** — Optically modulated and optically activated delayed fluorescent proteins through dark state engineering J. Phys. Chem. B 125 (20) · 2021

← The spectrum Retention projection →

B. Peng, Ph.D. — AppCubic · 2026 Imagery : Wellcome Collection CC-BY · RCSB PDB CC0 · computed in-house Google Scholar ↗