PI-Mamba: Linear-Time Protein Backbone Generation via Spectrally Initialized Flow Matching
This addresses the trade-off between efficiency and structural fidelity in protein backbone design, offering a scalable solution for computational biology.
The paper tackled the problem of generating protein backbones with geometric validity and computational efficiency, achieving 0.0% local geometry violations and high designability (scTM = 0.91±0.03) while scaling to over 2,000 residues on a single GPU.
Motivation: Generative models for protein backbone design have to simultaneously ensure geometric validity, sampling efficiency, and scalability to long sequences. However, most existing approaches rely on iterative refinement, quadratic attention mechanisms, or post-hoc geometry correction, leading to a persistent trade-off between computational efficiency and structural fidelity. Results: We present Physics-Informed Mamba (PI-Mamba), a generative model that enforces exact local covalent geometry by construction while enabling linear-time inference. PI-Mamba integrates a differentiable constraint-enforcement operator into a flow-matching framework and couples it with a Mamba-based state-space architecture. To improve optimisation stability and backbone realism, we introduce a spectral initialization derived from the Rouse polymer model and an auxiliary cis-proline awareness head. Across benchmark tasks, PI-Mamba achieves 0.0\% local geometry violations and high designability (scTM = $0.91\pm 0.03$, n = 100), while scaling to proteins exceeding 2,000 residues on a single A5000 GPU (24 GB).