SEEKER: CRISPR-powered Quantitative Keyword Search Engine in DNA Data Storage. Raw reads of 40 abstracts encoded using NCG coding

Archiving information in synthetic DNA has emerged as an attractive solution to deal with the exploding growth of data in the modern world. Random access to data stored in DNA has been achieved through file names, yet a convenient and direct way of quantitatively looking up the exact content of data stored in DNA sequences is still under investigation. Here, we developed Search Enabled by Enzymatic Keyword Recognition (SEEKER), which utilizes CRISPR-Cas12a to generate a visible fluorescence response when a DNA target corresponding to a particular piece of information is present. SEEKER can be applied as a search tool to determine the presence of a keyword in selectively amplified DNA text files. The growth rate of fluorescence intensity is proportional to the number of times the keyword appears, making SEEKER a biochemical computer capable of performing quantitative text searches. Compatible with SEEKER, we developed non-collision grouping (NCG) coding to encode and losslessly compress text files without disrupting the original order of texts. NCG coding is a fixed-length encoding technique allowing searches performed using an invariant length of query sequence, ideal for CRISPR-based detection. The dictionary generated through NCG coding can be stored in parallel with the text data and occupies just a small portion of the oligo pool. Query sequences can be determined after sequencing the reference strands working as the dictionary, rather than retrieving the entire dataset which is necessary for conventional compression techniques like Lempel–Ziv–Welch (LZW) coding. Both the text data and dictionary can be comprehensively recovered at a regular sequencing coverage. SEEKER can also be miniaturized on a 3D-printed microfluidic chip to allow easier operation. Overall, SEEKER provides a quantitative approach to conducting parallel searching over the complete content stored in DNA with simple implementation and rapid result generation.