SemiSynBio-II: DNA-Based Memory for High-Density Information Storage and Molecular Cryptography with Fast Readout Methods

随着时间的推移，对安全的计算机和互联网存储的需求迅速增加。为了应对这一挑战，该项目将开发一种基于DNA纳米技术的密码协议，用于写入和读取数字数据。它的开发将使高容量和高度安全的信息存储和高速信息检索成为可能。DNA纳米技术将被用来加密DNA折叠形成的分子模式中的信息，DNA折叠是由一组DNA信息链和作为加密密钥的DNA链的特殊折叠路线唯一确定的。DNA纳米结构的可编程性为加密密钥创造了巨大的设计空间，使得拦截和破解加密几乎是不可能的。DNA纳米结构上的分子图案将通过超分辨率成像方法和低噪声纳米孔设备来表征。这些方法可以实现1兆赫的快速读出速度和小于10纳米的高空间分辨率。如果成功，这种基于DNA的存储有可能在便携式平台上实施，从而有力地支持以低成本显著提高安全性和存储密度的下一代基于DNA的存储器的开发。因此，基于DNA的安全存储将使美国经济和社会受益。这项研究涉及多个学科，包括纳米制造、材料科学、生物化学、电子学、光子学和数据科学。该项目将使未被充分代表的群体参与研究，并对STEM教育产生积极影响。DNA由于其高稳定性、高信息密度和伴随的读出技术，已成为存储可与半导体存储器相媲美的信息的有前途的候选者。然而，DNA作为一种高度安全的信息存储材料的潜力尚未得到充分开发。此外，传统的基于DNA的记忆的读出受到DNA测序的限制，这需要特殊的仪器和人员培训。这项研究旨在通过创建一种显著改进的、高安全性的基于DNA的分子密码学来填补这一知识空白，并建立一个高速、高分辨率、潜在的便携式DNA存储读出平台。研究小组将以纳米图案的形式将加密信息嵌入三维DNA折纸纳米结构中。将根据从多种可能性中选择的DNA折叠方案来设计装饰有不同序列、长度和结合位置的特定信息链(I-链)的DNA支架链池，该方案将在稍后解密I-链的图案时使用。DNA-PAINT(用于纳米地形成像的基于DNA的点积累)(20 Nm)和蓝宝石支持的纳米孔传感器(10 Nm)的高空间分辨率将支持DNA折纸上复杂单分子图案的高密度信息解密，同时实现高达1 MHz的快速读出速度。快速读出方法将使用深度学习分类技术来实现自动解密和提高精度。在这个项目中，加密方案、单分子表征方法、纳米孔传感器设计和数据分析算法将得到协同改进，以实现强大而有效的信息存储和传输。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The need for secured computer and internet storage has seen a rapid increase over time. To address this challenge, this project will develop a DNA nanotechnology-based cryptographic protocol for writing and reading digital data. It will be developed to enable both high-capacity and highly-secured information storage and high-speed information retrieval. DNA nanotechnology will be utilized to encrypt information in molecular patterns formed upon DNA folding, which is uniquely determined by a set of DNA information strands and a special folding route of DNA strands that function as the encryption keys. The programmability of DNA nanostructures creates a vast design space for the encryption keys, making it practically impossible to intercept and break the encryption. The molecular patterns on DNA nanostructures will be characterized by a super-resolution imaging method and a low-noise nanopore device. These methods can enable fast readout speed at 1 megahertz with a high spatial resolution of smaller than 10 nanometers. If successful, this DNA-based storage has a potential to be implemented on a portable platform, thus strongly supporting the development of next-generation DNA-based memories with significantly improved security and storage density at a low cost. Consequently, the secured DNA-based storage will benefit the U.S. economy and society. This research involves several disciplines including nanofabrication, materials science, biochemistry, electronics, photonics, and data science. The project will engage underrepresented groups in the research and positively impact STEM education.DNA has emerged as a promising candidate to store information that can rival semiconductor memory, due to its high stability, high information density, and accompanying readout technologies. However, the potential of DNA as a material for highly-secured information storage has not been fully exploited. Additionally, the readout of conventional DNA-based memories has been limited by DNA sequencing, which requires special instruments and personnel training. This research is to fill the knowledge gap by creating a significantly improved and high-security DNA-based molecular cryptography and establishing a high-speed, high-resolution, and potentially portable readout platform for DNA memory. The research team will embed encrypted information in three-dimensional DNA origami nanostructures in the form of nanoscopic patterns. A pool of DNA scaffold strands decorated with specific information strands (i-strands) of different sequences, lengths, and binding positions will be designed on the basis of a chosen DNA folding scheme out of many possibilities, which will be used later in decryption of the pattern of i-strands. The high spatial resolution of DNA-PAINT (DNA-based point accumulation for imaging in nanoscopic topography) (20 nm) and sapphire-supported nanopore sensors (10 nm) will support high-density information decryption of intricate single-molecule patterns on the DNA origami, while enabling a fast readout speed of up to 1 MHz. The fast readout methods will employ deep-learning classification techniques for automated decryption and improved accuracy. The encryption scheme, single-molecule characterization methods, nanopore sensor designs, and data analysis algorithms will be synergistically improved in this project to enable robust and effective information storage and transmission.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.