Time-shifts provide high precision measurement of growth rate Since the relation τ = (1/μ max) ln (X2/X1) governs the time shift (τ) between ABT-737 nmr different growth curves, τ can be plotted as a function of ln (X2/X1) yielding a straight line with a slope of 1/μ max. This allows calculating the maximum specific growth rate (μ max) from the growth curve synchronization. When performing this quantification, we observed that WT and NEG have comparable growth rates (Figures 3 and 5A; μ max = 0.29 ± 0.02 h-1 versus μmax = 0.28 ± 0.01 h-1, respectively), which was already shown qualitatively

in previous experiments with rich media based on casamino acids and in direct competition experiments [13]. QSN also showed growth rates comparable to WT in the absence of C4-HSL (Figure 5B, squares; μ max = 0.27 ± 0.01 h-1). However, when C4-HSL was added to the media, QSN grew markedly slower (Figure 5B, triangles; μ max = 0.22 ± 0.02 h-1). C4-HSL was solubilized in acetonitrile, but the addition

of acetonitrile without autoinducer did not affect growth (data not shown). To the best of our knowledge, this effect has not been observed before. The addition of 0.5% L-arabinose to the growth media of IND did not affect their growth, as the growth rate was similar to WT cells (Figure 5C; μ max = 0.27 ± 0.01 h-1). Discussion We introduced the method of growth curve synchronization for the a posteriori synchronization of high-resolution time series and integration of online spectrophotometric data with endpoint Selleck Talazoparib measurements. We demonstrated the method with growth curve data from the opportunistic human pathogen Pseudomonas aeruginosa PA14 and isogenic mutants. The quality of the growth-curve SPTLC1 alignments was assessed by measuring the R2-values

for the linear regression of the calculated time-shift (τ) versus the logarithm the inoculum (R2 > 0.99 in all cases, Figures 3 and 5), a relationship that we formulated based on a simple mathematical model of exponentially growing cell cultures. In addition to carrying out data integration, our method provides a high-precision measurement of maximum specific growth rate. Figures 3 and 5 show the maximum specific growth rates (μmax) measured from the slope of the τ vs. ln(X2/X1). The average error of these measurements evaluated from the regression was 5.4%. In the worst case, being QSN in the presence of C4-HSL (Figure 5B, triangles), the error was 9.1%. This precision is quite good for growth rates measured from optical density, approaching the 5% error reported for a high-precision bioluminescence-based method [36]. However, in contrast to a bioluminescence assay, our OD-based method does not require introduction of a constitutively expressed luciferase reporter or the use of an expensive bioluminescence-capable reader.