Latent space mapping of interpretable structural coordinates from stochastic single-molecule signals

2026-06-15

AI Summary

This work addresses the challenge of temporal signal distortion in nanopore single-molecule sensing, which arises from stochastic translocation dynamics and impedes accurate decoding of molecular structural information. To overcome this limitation, the authors propose a physics-informed contrastive learning framework that leverages physically grounded signal simulations to train an encoder, enabling direct mapping of raw ionic current signals into an interpretable latent space aligned with molecular structural coordinates. This approach achieves, for the first time, an end-to-end transformation from time-domain signals to a structure-aware latent representation, facilitating cross-device data integration and real-time molecular identification via a single forward inference pass. The method reduces computational cost by three orders of magnitude while demonstrating high efficiency and robustness in applications including quantitative analysis of DNA barcode mixtures, rare variant detection, and real-time signal processing.

Abstract

Nanopores are versatile single-molecular sensors, but their utility is fundamentally constrained by stochastic translocation dynamics warping any encoded information. We resolve it by shifting from time-domain analysis to a learned latent-space mapping via a contrastive encoder trained exclusively on simulated signals from a physics-informed model. This encoder maps solid-state nanopore signals of engineered DNA barcodes into an interpretable molecular coordinate system. The learned representation is responsive to structural barcode parameters while remaining invariant to acquisition conditions and translocation conformation, allowing data pooling across devices. Molecule identification requires a single pass through the encoder, reducing computational cost by three orders of magnitude relative to alignment-based methods. We experimentally validate through mixture quantification, rare-variant detection, consensus barcode reconstruction, and real-time signal acquisition. This shift from temporal analysis to mapping structural coordinates into a latent space changes the paradigm behind analyzing stochastic sensor signals by linking classification to interpretable encoded molecular information.

Problem

• nanopore
• stochastic translocation
• single-molecule sensing
• signal distortion
• molecular structure decoding

Innovation

• latent space mapping
• contrastive encoder
• nanopore sensing
• interpretable representation
• stochastic signal analysis