Skip to main content
Fig. 8 | BMC Bioinformatics

Fig. 8

From: EFMlrs: a Python package for elementary flux mode enumeration via lexicographic reverse search

Fig. 8

Visualization of the scaling behavior of the sub-models of JCVI-syn3A [31]. The plot shows the correlation between the included number of non-essential uptake reactions, the number of resulting EFMs, and the run times. The sub-models are plotted on the x-axis and referred to by their number e.g. 00 refers to \(sm_{00}\). Y-axis 1 shows the number of calculated/predicted EFMs, y-axis 2 the required/predicted run time in days using 960 threads on the VSC. Both axes are in logarithmic scale. Full symbols (dots and squares) refer to measured run times and computed EFMs (sub-models \(sm_{00}\) to \(sm_{15}\)). The corresponding equations for the non-logarithmic representation of the calculated results are \(y1=264629e^{0.6678x}\) with \(R^2=0.9997\) and \(y2=3E^{-5}e^{0.7508x}\) with \(R^2=0.9986\). Empty symbols refer to estimated data (sub-models \(sm_{16}\) to \(sm_{22}\)) with the assumption that the slopes for both curves remain the same over all sub-models. Sub-model \(sm_{00}\) refers to the smallest sub-model that includes none of the 22 non-essential uptake reactions and \(sm_{22}\) to the full model that includes all uptake reactions

Back to article page