FiatFlux-P - User-Level Workflow Design: A Bioinformatics Perspective - page 108

Fig. 5.6 FiatFlux worfklow for the analysis of a single data set
5.2.3 Batch Processing FiatFlux Workflows
Fig. 5.7 METAFoR and 13 C-based metabolic flux analysis of several data sets
The actual objective of the FiatFlux-P project was to automate the Fiat-
Flux analysis procedure in order to obtain a standardized analysis process and
to increase the amount of data sets that can be analyzed in a certain amount
of time, that is, to implement a workflow that allows for batch processing of
numerous .cdf data sets. In FiatFlux, all experimental data have to be en-
tered manually by the user at different steps of the analysis procedure and at
different parts in the GUI. Figure 5.7 shows a workflow for the processing of
several data sets: The user has to specify the working directory and a .csv file
that lists the .cdf files under consideration and all experimental parameters.
(The .csv format can be exported from all common spreadsheet programs,
thus researchers can continue to document their experiments within MS Ex-
cel, OpenOce Calc or other.) The batch processing workflow then simply
repeats the single data set analysis workflow described above, processing an-
other data set in each iteration, until all input files have been analyzed. As
user input is only required once at the beginning, this workflow is able to
process very large sets of input data autonomously, speeding up the analysis
procedure significantly.
5.2.4 Further Variations of the FiatFlux-P Workflow
Instead of the standard Netto service that has been used in the examples
above, the Netto CustomModel and Netto JointRatios variants can be in-
cluded in the analysis workflows. For the former, it is additionally required to
provide a custom network model file, whereas for the latter a second ratio