SemiSynBio: Nucleic Acid Memory

云计算的快速扩散和海量科学、金融、政府和基因记录的大数据的出现正在造成信息存储危机。这些数据一旦产生，就会贯穿信息存储生命周期--从硬盘和固态驱动器形式的主存储介质到磁带等存档介质。虽然在信息密度、稳定性和能源消耗方面的创新屡见不鲜，但现有的存储材料正在接近其物理和经济上的终点。正如半导体合成生物学(SemiSynBio)路线图所设想的那样，基于DNA的海量信息存储是存储器制造的一个全新的开始。因此，研究团队提议通过设计、构建和测试两种由DNA制成的可访问、可编辑和非易失性存储技术来开创冷存储范例。受DNA电路的启发和最先进的光学物理学的启发，该团队将：(1)生物合成DNA分子，(2)设计由所述分子制成的基板，(3)使用额外的DNA分子将数字信息写入基板，(4)使用计算机科学算法将编码和解码错误降至最低，以及(5)使用可逆的DNA绑定在基板上读取和编辑数字信息。在这一跨学科项目的全力支持下，研究团队包括：DNA纳米技术、纳米尺度表征、光学物理、生物启发的算法和合成生物学方面的专业知识。以教员合作为蓝本，一批新的学生将在生物、计算和工程科学的交汇处工作和学习，期待着一个名为核酸记忆(NAM)的新兴领域。作为一个名为NAM的垂直整合项目的积极参与者和共同所有者，本科生和研究生将加入一个多年和多学科的研究团队，提供持续的课程和教学学分。这项建议的焦点是两个存储介质原型，数字核酸(DNaM)和序列核酸存储器(SeqNAM)。它们都提供了一种使用DNA对信息进行编码的新方法，并且都使用超分辨率显微镜来读取信息。在dNaM中，信息被编码成可寻址的DNA折纸纳米结构(称为NAM存储节点)上的DNA序列的特定空间排列。DNA折纸为NAM结节结构的高产量和快速成型提供了一种方便的途径和经过验证的方法。钉钉链将从NAM节点结构延伸出来，具有独特的序列，用于特定地点附着NAM数据链。当绑定时，数据链充当互补数据成像器链的对接位置，这些数据成像器链被用于基于DNA的超分辨率显微镜(SRM)形式，称为DNA-PAINT。DNA Paint是一种随机超分辨率成像技术，它使用荧光标记的数据成像器链的重复、瞬时结合来绕过光的衍射限制。因此，数据成像器串充当读取头，并以优于7 nm的分辨率显示NAM存储节点的每个站点的状态。每个数据单元处的二进制状态可以由由SRM确定的NAM数据链的存在(1)或不存在(0)来定义。通过多个正交序列可以简单地将特定于站点的比特密度从1比特增加到3比特。通过将所需的数据串添加到空闲的数据单元或通过脚尖调节的串置换移除现有的数据串来执行数据串的编辑。SeqNAM构建在类似的存储节点平台上，它使用两个数据单元将数据串排列成有序的阵列。在seqNAM中，信息被编码在保持单链的数据链部分内。使用多色超分辨率测序(SRS)过程读取数据链的序列，该过程使用锁定的核酸成像器链库。编辑是通过使用Toehold介导链入侵去除具有互补序列的目标数据链，然后添加替换数据链来执行的。SeqNAM通过在DNA序列中以潜在更高的密度存储信息而超过dNaM。此外，它还创建了一个新的无酶测序平台。这一奖项反映了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 rapid proliferation of cloud computing and the emergence of big data for massive scientific, financial, governmental, and genetic records is creating an information storage crisis. These data, once generated, cascade through the information storage lifecycle -- from primary storage media in the form of hard disks and solid-state drives to archival media such as magnetic tape. While innovations in information density, stability, and energy consumption routinely occur, existing memory materials are approaching their physical and economic finish lines. As imagined by the Semiconductor Synthetic Biology (SemiSynBio) Roadmap, DNA-based massive information storage is a brand new start for memory manufacturing. As a result, the research team proposes to pioneer a cold storage paradigm by designing, building, and testing two accessible, editable, and non-volatile memory technologies made from DNA. Inspired by DNA circuits and made possible by state-of-the-art optical physics, the team will: (1) biologically synthesize DNA molecules, (2) engineer substrates made from said molecules, (3) write digital information onto the substrates using additional DNA molecules, (4) minimize encoding and decoding errors using computer science algorithms, and (5) read, as well as edit digital information onto the substrates using reversible DNA binding. In full support of this interdisciplinary project, the research team includes expertise in: DNA nanotechnology, nanoscale characterization, optical physics, biologically-inspired algorithms, and synthetic biology. Modeled after the faculty collaboration, a new cadre of students will work and study at the confluence of the biological, computational, and engineering sciences in anticipation of the emerging field called Nucleic Acid Memory (NAM). As active participants in and co-owners of a Vertically Integrated Project called NAM, undergraduate and graduate students will enroll into a multi-year and multi-disciplinary research team that provides ongoing course and teaching credit.The focal points of this proposal are two storage medium prototypes, digital Nucleic Acid (dNAM) and sequence Nucleic Acid Memory (seqNAM). Each offer a novel approach to coding information using DNA, and both use super-resolution microscopy to read information. In dNAM, information is encoded into defined spatial arrangements of DNA sequences on top of addressable DNA origami nanostructures, called NAM storage nodes. DNA origami provides a convenient pathway and a proven approach to high-yield and rapid prototyping of NAM node structures. Staple strands will be extended from the NAM node structures with a unique sequence for site-specific attachment of NAM data strands. When bound, data strands serve as docking sites for complementary data imager strands, which are employed in a DNA-based form of super-resolution microscopy (SRM) called DNA-PAINT. DNA PAINT is a stochastic super-resolution imaging technique that uses repetitive, transient binding of fluorescently labeled data imager strands to circumvent the diffraction limit of light. Thus, data imager strands act as the read head and reveal the state of each site of the NAM storage node with better than 7 nm resolution. Binary states at each data cell can be defined by the presence (1) or absence (0) of the NAM data strand, as determined by SRM. Increasing site-specific bit-density from 1 to 3 bits can be simply achieved by multiple orthogonal sequences. Editing of data strands is performed by either adding a required data strand to a vacant data cell or by removing an existing data strand via toehold-mediated strand displacement. Built upon a similar storage node platform, seqNAM employs two data cells to arrange data strands into ordered arrays. In seqNAM, information is encoded within portions of the data strands that remain single stranded. The sequences of the data strands are read using a multi-color super-resolution sequencing (SRS) process that uses a library of locked nucleic acid imager strands. Editing is performed by removing the target data strands with complementary sequences using toehold-mediated strand invasion and then adding the replacement data strands. seqNAM exceeds dNAM by storing information within DNA sequences at a potentially higher density. In addition, it creates a new enzyme-free sequencing platform.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.