Quantum-driven Path Exploration Algorithm: QAOA for de novo Genome Assembly
Year: 2026
Type: preprint
Abstract: Metagenomic analysis of the 'microbial dark matter'-the vast majority of microorganisms lacking reference genomes-depends critically on de novo assembly methods. These approaches enable genome reconstruction directly from sequencing data without prior genomic knowledge, revealing previously unknown microbial diversity. Here, we introduce a graph transformation algorithm that converts bidirected compressed De Bruijn genomic graphs into directed graphs, then formulate the de novo assembly challenge as a Quadratic Unconstrained Binary Optimization (QUBO) problem addressable through the Quantum Approximate Optimization Algorithm (QAOA). Computational exploration was conducted on the Eviden Qaptiva 800 emulator using real-world viral data. Our analysis reveals that while QAOA successfully identifies valid assembly paths in small-scale instances, qubit requirements scale linearly with graph edge counts, presenting a bottleneck for larger bacterial genomes. To address this scalability issue, we propose a graph decomposition strategy to break down large assembly graphs into manageable subgraphs. This study establishes a proof-of-concept for quantum-enhanced de novo assembly, exploring the capabilities of QAOA against classical solvers and highlighting a path toward quantum-driven de novo assembly.
Source: HAL
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7159556033
