Codebox 1

Pseudocode for automated calculation of growth rates, using backscatter (BS) signal as example biomass signal. The procedure starts with the calculation of the blank (zero) value from first BS signals, where the cell concentration is below the limit of detection LOD (lines 1–3). These first BS signals below the LOD are also used to calculate the BS measurement error which is considered to be additive for all BS signals (lines 4, 5). Next, the measurement cycle is identified where the BS signal exceeds the user defined limit of quantification LOQ (lines 6–13). To detect the exponential growth phase from the complete BS data set, a BS subset is extracted ranging from measurement cycles where the BS signal reaches the LOQ to the last one. For this BS subset, the growth rate is calculated by a weighted linear regression and several stopping criteria are evaluated. If these criteria are not fulfilled, a new BS subset is evaluated from which the last measurement is removed. This procedure is repeated until the stopping criteria are fulfilled. Please note: this is implemented as for-loop in the MATLAB function (lines 14–28). Three stopping criteria are defined: A certain adjusted R² from the regression has to be reached. The last biomass increase has to be higher than the previous one, and these two increases need to have a positive sign (lines 29–34)