As these examples
show, autonomous 13C flux analysis—as any automation—entails the risk that raw data of insufficient quality are processed. Therefore, the implementation of routines SB202190 in vivo checking the quality of the original data, e.g. checking for detector overload and data of signals of insufficient intensity are crucial. In Flux-P, MDVs are removed from the analysis, if they cause improper flux ratios assuming a faulty MDV value of this particular fragment. However, equally possible is the use of incomplete or erroneous metabolic networks used for the flux ratio calculation. In order to prevent potentially Inhibitors,research,lifescience,medical wrong MDV exclusions and disclosing faulty networks, routines that check alternative network models have to be implemented. In summary, the automated analysis of 13C labeling data Inhibitors,research,lifescience,medical with Flux-P allows not only a fast pre-screening or initial analysis of large amounts of data but the fully automated calculation of high quality metabolic flux ratios and intracellular fluxes.
Observed differences from manually calculated flux distributions can be attributed to shortcomings of Inhibitors,research,lifescience,medical the analyzed data to unambiguously resolve all metabolic fluxes rather than to errors in the automated calculation. 3. Conclusions Existing software for 13C-based metabolic flux analysis—such as FiatFlux, OpenFLUX or 13CFLUX—supports experts in the complex analysis of intracellular fluxes, but requires several steps that have to be carried out manually, hence restricting their use for data interpretation to rather small numbers of experiments. Flux-P makes
it possible to automatically process 13C-based MFA of single as well as numerous input data sets. The interactive steps that are essential in the Inhibitors,research,lifescience,medical underlying software (FiatFlux in the current prototypical implementation) are replaced by specific scripts that emulate the user interaction, owing to the observation that the user acts, to a considerable extent, according to quantifiable criteria. In addition Inhibitors,research,lifescience,medical to the significant acceleration of the analysis process, Flux-P achieves a consistent analysis workflow and applies the same set of parameters to each data set, directly producing comparable all results. We showed that it is easy to integrate software as services via the jETI technology as soon as it can be operated in headless mode. The functions of the software are then available as platform-independent services and can be used for agile workflow definition within Bio-jETI. Encouraged by the good results that we have obtained with the prototypic implementation described in this paper, we are going to follow the approach further. Next to the implementation of data quality and model validity checks, discussed in section 2.8, we envisage the implementation of the analytic framework presented by Rantanen et al.