nanopore

nanopore integrates a Viterbi error-correction decoder with deep learning-based basecalling and convolutional coding to improve sequence accuracy for DNA-based data storage using nanopore sequencing.

Key Features:

• Integration of Viterbi decoder: Integrates a Viterbi error correction decoder directly with the basecalling process to utilize soft information from deep learning-based basecallers and improve sequence accuracy.
• Use of convolutional codes: Employs convolutional coding for error correction, reducing reading costs by approximately threefold while maintaining comparable writing costs.
• High-throughput nanopore compatibility: Targets nanopore sequencing technology, which enables high-throughput sequencing using compact devices suitable for DNA storage applications.

Scientific Applications:

• DNA-based data storage: Enables encoding and retrieval of data in DNA with high storage density (petabytes per gram) and long-term durability (lasting thousands of years).
• Reliable nanopore read retrieval: Addresses high error rates in nanopore reads by improving error correction to make reading stored DNA more reliable and cost-effective through reduced reading costs.

Methodology:

Integrates a Viterbi decoder with the basecalling process by leveraging soft outputs from deep learning-based basecallers and applies convolutional coding techniques for error correction.