Table 2. Model parameters values.
| Parameter | Description | Value |
|---|---|---|
| kbi1 | S+C rate constant; rxn (1) and (3); fitted | 5.5 × 105 M−1 s−1 |
| kbi2 | F+Int rate constant; rxn (1) and (2); fitted | 6.2 × 106 M−1 s−1 |
| kuni | Int2→C+B2 rate constant; rxn (3); fitted | 4.2 × 10−2 s−1 |
| kleak | F+S leakage rate constant; rxn (4); fitted | 7.4 M−1 s−1 |
| ksink | D+sink rate constant; rxn (5); fitted | 105 M−1 s−1 |
| kdeprot | D+PT rate constant; rxn (6); fitted | 107 M−1 s−1 |
| n | Phenomenological stoichiometry of nucleation; fitted | 2.5 |
| knuc | n ˙ monomer nucleation rate constant; rxn (7); fitted | 2.0 × 105 M1−n s−1 |
| kelong | Nanotube+Monomer rate constant; rxn (8); assumed | 3.4 × 106 M−1 s−1 |
The parameters of the catalysis subsystem were fitted only to the catalytic data shown in Supplementary Fig. S3, which includes fluorescence reporter characterization, the data shown in Figure 1b with 10 nM precursors, and catalytic behavior at 100 nM precursor concentrations (see Supplementary Methods for fitting details). We initially attempted to use more simplified reaction models for the catalysis subsystem, but these models failed to qualitatively capture its kinetic behavior over the relevant range of times and concentrations. Specifically, reaction (3) is essential for capturing both the unimolecular rate-limiting nature of the catalytic reaction at high substrate concentrations, and the product inhibition (due to reverse reaction) that slows the catalysis as the reaction approaches completion. Reactions (2) and (4) are modeled as irreversible, whereas in reality some amount of reverse reaction may occur (see Supplementary Note 3). A similar model was previously explored in the study by Zhang et al.48, and rate parameters reported in that work are all within one order of magnitude of corresponding rate constants here (with differences likely attributed to differences in DNA sequence).