Figure 1.
DNA polymerase (DNAP) as a molecular signal recorder.
(A) Overview of a strategy for using DNA polymerases as signal recording devices. Signals (top) are coupled to intracellular or extracellular cation concentration through direct or indirect modulation of an ion channel activity. Cation concentration is in turn coupled to DNA polymerase fidelity on a known template according to a known transfer function (orange curve), generating a DNA recording, in which data is represented by the density of misincorporated bases, and which can be read by DNA sequencing (bottom). (B) Modulation of Taq polymerase by Ca2+ concentration, measured by a traditional blue-white colony counting assay [25]. (C) Biochemical steps of the multiplex deep sequencing assay for measuring the transfer functions of error-prone DNAPs.
Figure 2.
Ion-dependent misincorporation rates of Dpo4 and Klenow exo− polymerases.
(A, B, C, D) Mean (top) and template-base-specific (bottom) misincorporation rates as a function of Mn2+ (A, C) and Mg2+ (B, D) concentrations. (E, F, G, H) Normalized distributions of misincorporated dNTPs for each template base. (I, J, K, L) Mean (top) and template-base-specific (bottom) misincorporation rates as a function of Ca2+ concentration at 200 µM background Mn2+ (I, K) and 7000 µM background Mg2+ (J, L) concentrations. Errors are given in Tables S1–2, and are shown as error bars in the line graphs when they are larger than the data symbol.
Figure 3.
Template position dependence of misincorporation rates.
(A) Template position dependence of Dpo4 misincorporation rates on the original template at varying Mn2+ (left) and Mg2+ concentration (right). (B) Template position dependence of Dpo4 misincorporation rates on the swapped template at varying Mn2+ (left) and Mg2+ concentration (right). (C) Template position dependence of Klenow exo− misincorporation rates on the original template at varying Mn2+ (left) and Mg2+ concentration (right). Letters above each data point denote the identity of the template base at that position. Grey shaded areas indicate the noise floor, defined as the maximum over positions of the misincorporation rate (plus SEM) observed in an identical experiment with Pfusion HF DNA polymerase (Figure S1). Red (blue) shaded areas in (A) and (B) correspond to shared sub-sequences between the original and the swapped template.
Figure 4.
Statistical analysis of misincorporation by Dpo4.
(A) Spatial dependence (un-normalized) of Dpo4 error rate at 800 µM Mn2+ on the original template (blue curve), and generalized linear model fits of this data set with respect to itself (green curve), and with respect to the swapped template data set (red curve). (B) Spatial dependence (un-normalized) of Dpo4 error rate at 800 µM Mn2+ on the swapped template (blue curve), and generalized linear model fits of this data set with respect to itself (green curve), and with respect to the original template data set (red curve). (C) Feature weights for generalized linear model fit to Dpo4 original template data. (D) Feature weights for generalized linear model fit to Dpo4 swapped template data. (E) Information gain per base as a function of template position, for discrimination between high (800 µM) and low (75 µM) Mn2+ by Dpo4. (F) Information gain per base as a function of template position, for discrimination between high (7000 µM) and low (1000 µM) Mg2+ by Dpo4.