Quantum Simulation of Protein Fragment Electronic Structure Using Moment-based Adaptive Variational Quantum Algorithms

Biraja Ghoshal · 2026

Read Quantum Simulation of Protein Fragment Electronic Structure Using Moment-based Adaptive Variational Quantum Algorithms by Biraja Ghoshal (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-00656.txt.

Introduction

Biraja Ghoshal is the sole author of a 2026 paper on quantum runs of bits of proteins. The aim is to learn how electrons act in the active sites of proteins. Ghoshal builds a hybrid frame: a quantum method does the hard core sums.

The test case is a small four-orbital bit of a serine protease, a protein that cuts other proteins. The method hit the mark set for chemical accuracy, a very tight error bar. It also won back about ninety-five in each hundred of the energy that plain methods miss. The author then tried the frame on wider cases. One case was a target tied to the virus that causes Covid-19. One was a liver protein that breaks down drugs, where it called most sites right. The work maps a path for quantum help in drug science.

Catalog record

CREATORBiraja Ghoshal
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE43 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-00656

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Quantum Simulation of Protein Fragment Electronic Structure Using Moment-based Adaptive Variational Quantum Algorithms Biraja Ghoshal Abstract Background: Understanding electronic interactions in protein active sites is fundamental to drug discovery and enzyme engineering, but remains computationally challenging due to exponential scaling of quantum mechanical calculations. Results: We present a quantum-classical hybrid framework for simulating protein fragment electronic structure using variational quantum algorithms. We construct fermionic Hamiltonians from experimentally determined protein structures, map them to qubits via Jordan-Wigner transformation, and optimize ground state energies using the Variational Quantum Eigensolver implemented in pure Python. For a 4-orbital serine protease fragment, we achieve chemical accuracy ( < 1.6 <1.6 mHartree) with 95.3% correlation energy recovery. Systematic analysis reveals three-phase convergence behaviour with exponential decay ( α = 0.95 \alpha=0.95 ), power law optimization ( γ = 1.21 \gamma=1.21 ), and asymptotic approach. Application to SARS-CoV-2 protease inhibition demonstrates predictive accuracy (MAE=0.25 kcal/mol), while cytochrome P450 metabolism predictions achieve 85% site accuracy. Conclusions: This work establishes a pathway for quantum-enhanced biomolecular simulations on near-term quantum hardware, bridging quantum algorithm development with practical biological applications. Keywords: quantum computing, variational quantum eigensolver, protein fragments, electronic structure, drug discovery, enzyme engineering Introduction Proteins are complex molecular machines whose function depends critically on electronic interactions within active sites and local fragments. The accurate simulation of electronic structure in biomolecular systems, particularly protein fragments, presents significant challenges due to strong electron correlation effects in enzymatic active sites containing transition metals, aromatic residue networks, or regions involved in charge transfer. Computational prediction of these interactions represents a central challenge in structural biology and drug discovery [ 30 ] . Computational methods that capture these interactions, including Hartree-Fock, Configuration Interaction (CI), and Density Functional Theory (DFT), become computationally intractable as system size increases due to exponential scaling of the Hilbert space [ 30 ] . Quantum computing offers a promising alternative through algorithms like the Variational Quantum Eigensolver (VQE), which provides a hybrid quantum-classical approach to approximate ground state energies for fermionic systems with polynomial scaling on quantum hardware [ 23 ] . Despite significant progress in quantum algorithms for chemistry, applications to biological systems remain limited. Most demonstrations focus on small molecules (H 2 , LiH, BeH 2 ), with protein fragments largely unexplored due to challenges in Hamiltonian construction, ansatz design, and biological interpretation [ 13 ] . Recent hardware advances, including IBM’s 127-qubit Eagle processor and Quantinuum’s high-fidelity trapped-ion systems, have enabled larger simulations, but biological applications remain nascent [ 26 , 24 ] . Here, we bridge this gap by developing a complete quantum-classical framework for protein fragment electronic structure calculations. We integrate experimental structural data from the Protein Data Bank with quantum algorithm optimization, achieving chemical accuracy for biologically relevant systems. Our approach addresses key challenges in biological quantum simulations, including Hamiltonian construction from real protein structures, development of protein-specific ansätze, and practical application validation against experimental data. Literature Review This review examines advances through 2025 in applying fermionic Hamiltonians and the Variational Quantum Eigensolver (VQE) to protein fragment calculations. The theoretical foundation for this work rests on the second-quantized fermionic Hamiltonian, which provides a precise description of interacting electrons in a molecular system [ 17 ] . For protein fragments, this Hamiltonian is constructed within a chosen basis set and often reduced to a manageable size through "active space" selection, where only the most chemically relevant orbitals are treated quantum-mechanically [ 7 ] . The core challenge lies in mapping this fermionic operator to a form executable on a quantum processor, using transformations like Jordan-Wigner or the more efficient Bravyi-Kitaev and parity encodings [ 21 , 1 ] . The VQE algorithm then addresses this Hamiltonian by using a parameterized quantum circuit, or ansatz, to prepare a trial wavefunction. A classical optimizer iteratively adjusts these parameters to minimize the energy expectation value. The design of an efficient, accurate ansatz—such as variations of the Unitary Coupled Cluster (UCC) ansatz or adaptive approaches—is critical [ 10 , 31 ] . Significant methodological advances through 2025 have been made to tailor these quantum algorithms for complex biological systems [ 12 ] . Research has focused on developing systematic, automated protocols for selecting chemically meaningful active spaces in metalloenzyme clusters, enabling more robust simulations with 20 to 40 qubits [ 7 ] . To combat the problem of deep quantum circuits, new ansatz structures like the iterative Qubit-Excitation-Based (QEB) ansatz and orbital-optimized VQE (OO-VQE) approaches have been developed and tested on biological fragments [ 34 , 28 ] . Furthermore, advanced techniques for grouping Hamiltonian terms (Clifford grouping) and error mitigation strategies (zero-noise extrapolation, probabilistic error cancellation) are now essential for improving accuracy on imperfect hardware [ 9 , 33 ] . Several computational studies up to 2025 demonstrate the potential and current limitations of this approach. Proof-of-concept simulations have targeted biologically crucial systems. For instance, [ 20 ] simulated a reduced 54-qubit model of the nitrogenase FeMo-cofactor, achieving improved spin-state energetics using error-mitigated VQE. Similarly, [ 3 ] investigated the oxygen-evolving complex of Photosystem II, highlighting the critical role of dynamical correlation not captured by VQE in small active spaces. Benchmarking studies on dipeptide models and tryptophan chains continue to validate methods against full configuration interaction, but underscore the precision gap [ 22 , 8 ] . Table 1 : Notable biological applications of quantum computing (2023-2025) System Method Qubits Accuracy Chlorophyll dimer VQE with ADAPT ansatz 18 2.1\text{\,}\mathrm{mE} HIV-1 protease active site Quantum embedding + VQE 14 3.5\text{\,}\mathrm{mE} Cytochrome P450 heme Error-mitigated VQE 16 1.8\text{\,}\mathrm{mE} Beta-lactamase inhibitor QM/MM with quantum core 12 4.2\text{\,}\mathrm{mE} Photosystem II Mn cluster Quantum Monte Carlo 20 5.1\text{\,}\mathrm{mE} Ribozyme catalytic pocket Fragment VQE 8 \times 4 fragments 2.9\text{\,}\mathrm{mE} • Note: Accuracy is reported relative to classical reference calculations (CCSD(T) or DMRG). Enzyme Active Sites: Several groups have reported quantum simulations of enzyme active sites: • Cytochrome P450 : Zhang et al. (2024) simulated the heme active site of cytochrome P450 using 16 qubits on IBM’s Heron processor, achieving chemical accuracy for the Fe-oxo bond energy [ 35 ] . Key innovations included a chemically informed basis set reduction and efficient error mitigation. • HIV-1 Protease : The Merck-IBM collaboration (2024) demonstrated quantum simulations of the HIV-1 protease active site using quantum embedding methods, achieving accuracy sufficient for ranking inhibitor binding affinities [ 11 ] . • Photosystem II : Preliminary simulations of the Mn 4 CaO 5 cluster in photosystem II have been performed using quantum Monte Carlo methods, though with limited accuracy due to the complex electronic structure [ 16 ] . Drug Discovery Applications: Quantum computing is beginning to impact drug discovery pipelines: • Binding Affinity Prediction : Bayer and Google (2023) reported quantum-enhanced predictions of protein-ligand binding affinities for kinase inhibitors, achieving correlation coefficients of 0.85 with experimental data [ 32 ] . • Metabolism Prediction : The AstraZeneca-Quantinuum collaboration (2024) developed quantum algorithms for predicting cytochrome P450 metabolism sites, achieving 80% accuracy on a test set of 50 drugs [ 4 ] . • Toxicity Prediction : Quantum machine learning models have been applied to predict cardiotoxicity (hERG channel inhibition) with AUCs up to 0.82 [ 6 ] . Despite promising progress, the field as of 2025 faces substantial challenges

Read the complete text