Scholarlink

<p>DNA inherently offers ultra-high storage density, exceptional longevity, and ultralow energy consumption, making it a transformative alternative to traditional semiconductor-based storage systems. Despite its potential, practical DNA storage faces critical bottlenecks, including low-throughput encoding, high error rates in molecular channels, and strict biological constraints. This study proposes a novel hybrid-encoding framework specifically designed for vector images—resolution-independent digital files retaining precision regardless of scaling. Our approach integrates: compressive hybrid encoding to maximize storage density while ensuring biological feasibility, error-resilient mechanisms (e.g., Reed–Solomon codes) mitigating DNA synthesis/sequencing errors, and biological constraint optimization by dynamically balancing GC content and homopolymer length. A visual interface tool automates bidirectional conversion between vector files and DNA sequences, enabling seamless storage, writing, and retrieval. Critically, we introduce contour-based associative retrieval, leveraging vector image topology to achieve similarity search across 100 images—an unprecedented feature in DNA storage systems. Performance evaluation through comprehensive simulations demonstrates: error reduction, precision retrieval and scalability. These advancements address long-standing challenges in random access, data editing, and scalability, positioning our system as a scalable solution for structured data storage in DNA.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

High-density DNA storage for vector images: hybrid encoding with error correction and contour-driven retrieval

  • 期刊论文

  • 发表日期：

    2025年8月4日

查看全文
  • Chunxia Ge,
  • Xiaosheng Dong,
  • Zhidong Xue

摘要

DNA inherently offers ultra-high storage density, exceptional longevity, and ultralow energy consumption, making it a transformative alternative to traditional semiconductor-based storage systems. Despite its potential, practical DNA storage faces critical bottlenecks, including low-throughput encoding, high error rates in molecular channels, and strict biological constraints. This study proposes a novel hybrid-encoding framework specifically designed for vector images—resolution-independent digital files retaining precision regardless of scaling. Our approach integrates: compressive hybrid encoding to maximize storage density while ensuring biological feasibility, error-resilient mechanisms (e.g., Reed–Solomon codes) mitigating DNA synthesis/sequencing errors, and biological constraint optimization by dynamically balancing GC content and homopolymer length. A visual interface tool automates bidirectional conversion between vector files and DNA sequences, enabling seamless storage, writing, and retrieval. Critically, we introduce contour-based associative retrieval, leveraging vector image topology to achieve similarity search across 100 images—an unprecedented feature in DNA storage systems. Performance evaluation through comprehensive simulations demonstrates: error reduction, precision retrieval and scalability. These advancements address long-standing challenges in random access, data editing, and scalability, positioning our system as a scalable solution for structured data storage in DNA.