Fig. 2: DORIS eliminates non-specific interactions and increases density and capacity limits.

a Single primer extension created ss-dsDNAs. (Bottom) 4 cycles of PCR generated the optimal amount of 160 nt ss-dsDNAs while minimizing excess ssDNA production. (Right) DNA gel showed a marked increase in generation of ss-dsDNAs below 1:10 ssDNA:primer ratios. b Individual files can be separated from a three-file database created by a one-pot single primer extension. Each file was bound by its corresponding biotin-linked oligo, followed by a non-PCR-based separation using functionalized magnetic beads. File separation specificity is the percentage of the DNA separated by that is either file A, B, or C as measured by qPCR. c (Left) PCR but not DORIS will allow oligos to bind internal off-target sites and produce undesired products. (Middle) DNA gels and (Right) their quantified fluorescence (blue for PCR, pink for DORIS) showed that PCR-based access resulted in truncated and undesired amplicons whereas DORIS accessed only the desired strands. d (Left) Monte Carlo simulations estimated the number of oligos found that will not interact with each other or the data payload. 400,000 oligos were tested against different density encodings. The x-axis represents density (Eq. (4)), which is inversely related to the length of codewords used to store discrete one-byte data values. We evaluated codeword lengths of 12 through 4. For DORIS, the encoding density was not impacted because it need not guard against undesired binding between the oligos and data payloads. (Right) For PCR, the number of oligos that will not bind the data payload drops as strand density increases, which means that fewer files can be stored, leading to a lower overall system capacity. For DORIS, the availability of oligos is independent of encoding, and capacity therefore increases with denser encodings. Plotted values represent the arithmetic mean, and error bars represent the s.d., of three replicate file separations or simulations. Gel images are representative of three independent experiments measured by RT-QPCR. Source data are provided as a Source Data file. *Capacities may be limited by synthesis and sequencing limitations not accounted for here.