AF: Medium: Collaborative Research: Top-down algorithmic design of structured nucleic acid assemblies

过去十年见证了使用自组装核酸“打印”复杂纳米级结构和图案的能力的显着增长。这些结构可用作合成 1-100 纳米级无机材料的模板，或直接用于基于 DNA 的记忆存储、治疗递送、单分子结构测定和纳米级激子材料等应用。虽然可以使用各种计算策略从底层 DNA 或 RNA 序列和拓扑结构手动设计这些复杂的 3D 结构，但仅从目标几何形状自主生成线性核酸序列的逆问题仍然是一个未解决的计算挑战。在该项目中，探索了全自动、自上而下的计算机辅助设计 (CAD) 算法，仅使用目标几何形状以自主方式为广泛的编程 DNA 和 RNA 组装体生成拓扑序列设计。这些组件可以通过体外或体内的自组装“打印”，以使用合成或转录的核酸形成目标纳米级几何形状。该方法将提供一种广泛使用的高级编程语言，以实现基于核酸的任意1D/2D/3D纳米级结构材料的基于序列的编程，在基础科学和纳米技术中具有多种应用。所提出的计算算法将作为开源软件在网上免费分发，并集成到各种软件包中，以广泛实现DNA和RNA组装的自上而下设计。这些算法和软件将为更广泛的科学和工业界提供易于使用的高水平设计策略，从而加速各团体广泛参与核酸纳米技术在生物分子和材料科学技术领域的各种应用。这些工具将为高中生和本科生提供获得核酸纳米结构设计实践经验的机会。亚利桑那州立大学和麻省理工学院的课程开发将采用该序列设计软件，让本科生和研究生参与其在计算机科学和纳米技术研究中基本问题的使用和应用。将探索使用 DNA 和 RNA 设计纳米级结构材料的基础方面。将研究使用 DNA 和 RNA 渲染各种基于 CAD 的几何基元的算法方法，包括 2D 和 3D 线框晶格、可能包含任意曲率的单层表面以及 3D 实体对象。网格划分算法将用于在 1D、2D 和 3D 中离散化几何对象，拓扑路由和序列设计将用于在 CAD 对象内定位核酸链。连续和不连续的单链核酸将使用反平行和平行交叉配置通过双链体，以利用不同的编程自组装模式。序列设计和路由将通过实验进行验证，以探索在 3D 空间中获得最佳折叠、自组装和特定碱基对定位的原理。此外，还将利用单个长连续支架链的无钉设计，探索 RNA 自组装纳米结构。实验和计算之间的密切相互作用将有助于提炼出编程结构化核酸组装体的基本而实用的方法。

The past decade has witnessed dramatic growth in ability to "print" complex nanometer-scale structures and patterns using self-assembling nucleic acids. These structures can be used as templates to synthesize inorganic materials on the 1-100 nanometer-scale, or employed directly in applications such as DNA-based memory storage, therapeutic delivery, single-molecule structure-determination, and nanoscale excitonic materials. While various computational strategies are available to forward design these complex 3D structures manually from underlying DNA or RNA sequence and topology, the inverse problem of autonomously generating linear nucleic acid sequences from target geometry alone remains an unsolved computational challenge. In this project, fully automatic, top-down computer-aided design (CAD) algorithms are explored to generate topological sequence designs for broad classes of programmed DNA and RNA assemblies in an autonomous manner using target geometry alone. These assemblies can be "printed" via self-assembly in vitro or in vivo to form target nanoscale geometries using either synthetic or transcribed nucleic acids. The approach will offer a broadly accessible, high-level programming language to realize sequence-based programming of arbitrary 1D/2D/3D nanoscale structured materials based on nucleic acids with diverse applications in basic science and nanotechnology.The proposed computational algorithms will be distributed freely online as open source software as well as integrated into a variety of software packages to broadly enable the top-down design of DNA and RNA assemblies. These algorithms and software will provide the broader scientific and industrial communities with easy-to-use, high-level design strategies that will accelerate the broad participation of groups in the use of nucleic acid nanotechnology for diverse applications in biomolecular and materials science and technology. The tools will open up opportunities for high school students and undergraduates to gain hands-on experience in nucleic acid nanostructure design. Curriculum developments at ASU and MIT will employ the use of this sequence design software for participation by undergraduate and graduate students in its use and application to basic questions in computer science and nanotechnology research.Foundational aspects of the design of nanoscale structured materials using DNA and RNA will be explored. Algorithmic approaches to rendering diverse CAD-based geometric primitives using DNA and RNA will be investigated, including wireframe lattices in 2D and 3D, single-layer surfaces that may contain arbitrary curvatures, as well as 3D solid objects. Meshing algorithms will be used to discretize geometric objects in 1D, 2D, and 3D, and topological routing and sequence design will be applied to position nucleic acid strands within CAD objects. Continuous and discontinuous single stranded nucleic acids will be routed through duplexes using anti-parallel and parallel crossover configurations to exploit distinct modes of programmed self-assembly. Sequence design and routing will be validated experimentally to explore principles for obtaining optimal folding, self-assembly, and positioning of specific base pairs in 3D space. Self-assembly of nanostructures from RNA will additionally be explored, utilizing staple-free designs from single long continuous scaffold strands. Close interaction between experiment and computation will help to distill fundamental yet practical approaches to programming structured nucleic acid assemblies.