SemiSynBio-II: Engineering Write, Access, Read, and Protect (WARP) Drives for DNA-based Data Storage Systems.

Digital data are being generated at increasingly overwhelming rates both economically and for the environment, with data storage material and energy usages close to reaching the physical limits of what current technologies can provide. For over two decades DNA has been considered with only limited enthusiasm as a potential next generation medium for data storage, despite holding pronounced advantages of high density, stability, and low energy requirements. This is because executing the core operations of a data storage system including write, access, read, and protect using DNA currently requires multiple complex steps carried out in succession. Furthermore, they often require specialized processing and purification in between each operational step. This leads to practical barriers in speed, cost, scalability, and reusability. The central hypothesis of this proposal is that an interdisciplinary collaboration can completely reimagine and reengineer these core operations to be continuous, highly practical, and economically viable through creative integration of knowledge and technologies from disparate disciplines. Specifically, molecular biology techniques from the Biological Sciences will be informed and designed by models and simulations derived from the discipline of Computer Science and Engineering, while these biological components will be physically controlled and interfaced with nanomaterials developed with expertise in Materials Science and Nanoscience. This hybrid information storage system engineered from DNA directly interfaced with novel nanofibrillar substrates, will leverage dynamic processing by biomolecular manipulations, and will unlock direct and rapid readout of information with semiconductor-based nanopore technology. Biomolecular-based computational modeling will inform all aspects of both the design of the system and its operation, from molecular to systems level scales. The interdisciplinary nature of this project will also provide strong opportunities to advance education and diversity in the Science-Technology-Engineering-Math (STEM) workforce through several avenues. This interdisciplinary research provides opportunities for undergraduate researchers to experience directly working with peers and PhD students from very different disciplines on common problems. Undergraduate course modules will be created that bridge synthetic biology, nanoscience, materials science, computer science and engineering that present applied interdisciplinary examples including DNA storage as case studies. Hands-on kits will also be developed to facilitate in-person and virtual activities administered through partnerships with organizations that reach and support underrepresented grade-school students in STEM.

This project aims to transform DNA storage systems from batch to continuous architectures. Each Aim addresses a unit process (WRITE, ACCESS, READ, PROTECT). WRITE: Rapid, scalable, and cheap DNA synthesis will be engineered by a combination of tree algorithms and enzyme-driven DNA assembly. ACCESS: Continuous and reusable access of information and files will be achieved by immobilization of a DNA database on dendritic colloidal particles with easily accessible high surface areas. This novel scaffold allows a transcription-based file access system previously developed by the team to rapidly extract information in the form of transcribed RNA through simple laminar flow through the DNA database. READ: RNA generated by the ACCESS technology will be directly read by nanopore sequencing, and fundamental experimental and machine learning studies will be engaged to inform the encoding algorithms of DNA databases and to optimize real-time base-calling. PROTECT: The unique physical and thermodynamic attributes of DNA and of these first three hybrid unit processes will be exploited to encrypt and obfuscate information. While fundamental thermodynamic findings, computational models, and physical technologies generated in each aim have strong and specific benefits in mutually informing the other aims, these unit processes will further be integrated together by two ongoing explicit Umbrella Tasks. End-to-end microfluidic devices storing protected data will be engineered by integrating all unit processes, informed by systems-level modeling to identify unit processes that are the bottlenecks for speed, cost, robustness, and longevity. This work will also generate important fundamental knowledge and models that will be generalizable to other types of biomolecular-based information storage systems as well as applicable to problems in each distinct discipline regarding DNA assembly and interaction specificity, nanoparticle engineering, and nucleic acid sequencing.

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.