How AI Reads DNA One Molecule at a Time ✎ express

Original: "Physics-Aligned Deep Learning Enables SERS Resolving and Sequencing of Dynamic Single-Molecule DNA Oligomers in Plasmonic Nanocavity"

· Kuo Zhan, Peilin Xin, Yingqi Zhao, Han Gu, Enock Adjei Agyekum, Zhou Chen, Shuai Li, Jian Ye, Lu Cheng, Jian-an Huang

arXiv:2608.16576v1 · 2026-08-17 · arXiv non-exclusive · independent analysis · 1 min

Optics

A neural network turns the jittery signals from a single DNA molecule into readable text.

A single DNA molecule, squeezed into a tiny metal slit, trembles and flickers under the laser. Its signal is not a clear string of letters but a chaotic mix of flashes and shadows. Deciphering this noise is like trying to make out a whisper in a crowded room.

The new neural network works like an attentive interpreter. It sifts through thousands of spectra, picking out those that show traces of individual nucleotides — the building blocks of DNA. Then it sorts them by type, creating a library of molecular states. From the transitions between states, the algorithm reconstructs the most likely sequence of the chain.

The transitions between states are like letters changing on a ticker tape: each step leaves a trace that lets you read the word.

In this way, the chaos of single molecules is turned into quantitative information. Already, the method can distinguish 12 states of DNA segments, and in the future it will allow reading any biopolymers without prior preparation.

Whole graph →

A single DNA molecule in a nanocavity produces tens of thousands of spectra per second — the neural network manages to pick out the most informative ones.

Scientists

Christian Doppler D. B. McLaughlin Didier Queloz Michel Mayor R. A. Rossiter James Clerk Maxwell

Concepts

sequencing PCR deep learning in medical imaging self-supervised learning deep neural network predictive model

I am the author

How AI Reads DNA One Molecule at a Time ✎ express