PhD Studentship: In Silico Directed Evolution of Biological Qubits: Rational Design of Nanoscale Quantum Transducers

Engineered flavoproteins, LOV domains, and fluorescent proteins have recently demonstrated that radical pair mechanisms (RPM) can serve as biological qubits with optical readouts, achieving magnetic field responses up to a 50% fluorescence change in living cells. However, reaching such a large magnetic sensitivity involved empiric and slow laboratory exploration, including directed evolution, and the (optimized) function is strictly that of a passive quantum observers. The scientific frontier of this project lies in the transition from passive quantum observers (fluorescence-based readouts) to active quantum transducers, i.e., genetically encoded systems that actuate a downstream biological response directly governed by quantum spin dynamics. Concurrently, rapid breakthroughs in generative AI and de novo protein design enable the systematic exploration of sequence and backbone space beyond natural scaffolds.

By unifying open quantum system spin dynamics simulations with deep-learning protein generation, this project will establish an in silico directed evolution pipeline to engineer biological qubits that transduce spin-dependent events into functional, biological effector signals.

Project Aims and Objectives

➢ Aim 1: Mapping Optimal Spin Dynamics to Structural Targets. Translate theoretically optimal spin Hamiltonians (maximizing magnetic field sensitivity via tuned hyperfine couplings, exchange interactions, and recombination kinetics) into concrete physical constraints, such as inter-cofactor geometries and local environments.

➢ Aim 2: In Silico Generative Protein Design. Develop a de novo design workflow leveraging deep-learning inverse-folding algorithms and electronic structure estimations to generate novel protein scaffolds and sidechain packings optimized for radical pair generation and (coherent) lifetime.

➢ Aim 3: Pipeline Calibration & Benchmarking. Calibrate the computational pipeline by reverse engineering existing laboratory directed-evolution constructs (such as MagLOV variants).

➢ Aim 4: Engineering Biological Transduction. Design allosteric coupling pathways wherein spin-selective chemical yields or radical lifetimes trigger structural shifts, converting the quantum sensor into a biological affector (e.g., modulating protein-protein interactions or enzymatic activity). Prospective predictions will be considered for experimental validation via Engineering and Physical Sciences Research Council Doctoral Landscape Award collaborating facilities at the Living Systems Institute (LSI) and National Physical Laboratory (NPL).

Methodology

The candidate will implement a multiscale computational framework combining open quantum spin dynamics (utilizing stochastic Liouville solvers and accounting for parametric uncertainty using a Bayesian approach) with generative AI protein design tools (e.g., ColabDesign, ProteinMPNN), molecular dynamics free-energy calculations, and ab initio prediction of spin Hamiltonian parameters and electron transfer rates (DFT based). The project is co-supervised by Prof. Alex Jones at NPL in support of the Laboratory’s strategic initiative in nanoscale quantum sensors for intracellular measurement and will leverage protein design methodology of start-up Orion Life Gmbh.

Funded by: EPSRC Engineering and Physical Sciences Research Council 

Advert information