SemiSynBio-III: Scalable Nucleic Acid Memory

全球数字数据生成的增长速度迅速超过了当前数据存储技术的容量。预计到2025年，每年将产生超过1500亿TB的数据。DNA由于其高稳定性和高信息密度而成为下一代数据存储解决方案。然而，由于DNA合成和测序所需的成本和能量，基于DNA的信息存储的潜力尚未得到充分开发。在这个项目中，自组装的DNA纳米结构在所有三个维度上以纳米精度放置DNA数据链。存储的信息通过时间分辨的超分辨率显微镜进行光学恢复。因此，对独特DNA链的需求显著减少，合成成本相应降低。此外，数据读出是非破坏性的，存储容量可扩展到TB级以上。该项目在合成生物学和半导体研究的界面上开发和提供劳动力发展培训。通过这个项目培训的学生将沉浸在一个跨学科的研究团队，在材料科学，合成生物学，计算机科学和电气工程的专业知识。研究结果将通过高影响力的期刊，总理会议，网站和社交媒体传播，并将被整合到多个课程在博伊西州。该研究团队将与博伊西州立包容和变革奖学金研究所合作，招募和留住来自代表性不足背景的学生。该项目旨在通过强大的数据编码/解码算法，基于深度神经网络的图像处理，可扩展的合成生物学，定制成像阵列和进化启发的序列优化算法来推进基于DNA的信息存储。为此，将开发一种旋转不变的数据编码、纠错方案，沿着最先进的基于深度神经网络的自适应图像处理方法，以及贝叶斯最优数据恢复/纠错算法，以优化探头定位和漂移校正。将创建具有正交DNA钉合链的大型定制DNA“支架”，以构建每次组装具有更多数据位点的折纸。将创建DNA支架，以允许在单个“一锅”合成中构建多个存储节点。将开发一种时间相关的超分辨率显微镜技术，该技术将超分辨率和荧光寿命显微镜无缝集成，以记录染料标记的成像链及其序列互补物之间的三维相互作用。基于最先进的单光子雪崩二极管技术的定制成像阵列将由商业代工服务设计和制造，以实现5 X 5 X 1 nm的读出分辨率。最后，将开发一种进化启发的序列优化算法，以优化DNA序列，用于读出和折纸合成。这项工作将提高DNA纳米结构中存储的信息密度和读取效率，从而减少大规模数据存储所需的DNA数量，提高基于DNA的信息存储的可扩展性和可持续性。该项目由工程局（ENG）的电气，通信和网络系统（ECCS）部门共同资助，计算机和信息科学与工程局（CISE）的计算和通信基础部（CCF），生物科学局（BIO）的分子和细胞生物科学部（MCB），以及数学和物理科学理事会（MPS）的材料研究部（DMR）该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Worldwide digital data generation is growing at a rate that rapidly outpaces the capacity of current data-storage technologies. Over 150 billion terabytes a year are predicted to be generated by 2025. DNA has emerged as a next-generation data storage solution due to its high stability and high information density. However, the potential of DNA-based information storage has not been fully exploited due to the cost and energy required for DNA synthesis and sequencing. In this project, self-assembled DNA nanostructures place DNA data strands with nanometer precision in all three dimensions. The stored information is recovered optically by time-resolved super-resolution microscopy. As a result, the need for unique DNA strands is significantly reduced with a corresponding decrease in the cost of synthesis. Furthermore, data readout is non-destructive and storage volume scales beyond terabyte levels. This project develops and delivers workforce development training at the interface of Synthetic Biology and Semiconductor research. Students trained through this project will be immersed in an interdisciplinary research team with expertise in Material Science, Synthetic Biology, Computer Science and Electrical Engineering. The research results will be disseminated through high-impact journals, premier conferences, websites, and social media, and will be integrated into multiple courses at Boise State. The research team will partner with the Boise State Institute for Inclusive and Transformative Scholarship to recruit and retain students from underrepresented backgrounds.The project aims to advance DNA-based information storage through robust data encoding/decoding algorithms, deep neural network-based image processing, scalable synthetic biology, custom imaging arrays, and an evolution-inspired sequence optimization algorithm. To do so, a rotation-invariant data-encoding, error-correction scheme will be developed along with a state-of-the-art deep neural network-based approach for adaptive image processing, and a Bayes-optimal data recovery/error correction algorithm to optimize probe localization and drift correction. Large custom DNA “scaffolds” with orthogonal DNA staple strands will be created to construct origami with more data sites per assembly. DNA scaffolds will be created to allow multiple storage nodes to be built in a single “one-pot” synthesis. A time-correlated super-resolution microscopy technique will be developed, which seamlessly integrates super-resolution and fluorescence lifetime microscopy to record interactions between dye-labeled imager strands and their sequence complement in three dimensions. Custom imaging arrays based on state-of-art single-photon avalanche diode technology will be designed and manufactured by commercial foundry services to achieve 5 X 5 X 1 nm readout resolution. Finally, an evolution-inspired sequence optimization algorithm will be developed to optimize DNA sequences both for readout and origami synthesis. In doing so, this work will increase information density that can be stored in DNA nanostructures and the efficiency of read, which reduces the amount of DNA needed for large scale data storage, improving both the scalability and sustainability of DNA-based information storage.The project was jointly funded by Division of Electrical, Communications and Cyber Systems (ECCS) in the Directorate for Engineering (ENG), Division of Computing and Communication Foundations (CCF) in the Directorate for Computer and Information Science and Engineering (CISE), Division of Molecular and Cellular Biosciences (MCB) in the Directorate for Biological Sciences (BIO), and the Division of Materials Research (DMR) in the Directorate for Mathematical and Physical Sciences (MPS).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.