PSAMM: A Portable System for the Analysis of Metabolic Models

Abstract

The genome-scale models of metabolic networks have been broadly applied in phenotype prediction, evolutionary reconstruction, community functional analysis, and metabolic engineering. Despite the development of tools that support individual steps along the modeling procedure, it is still difficult to associate mathematical simulation results with the annotation and biological interpretation of metabolic models. In order to solve this problem, here we developed a Portable System for the Analysis of Metabolic Models (PSAMM), a new open-source software package that supports the integration of heterogeneous metadata in model annotations and provides a user-friendly interface for the analysis of metabolic models. PSAMM is independent of paid software environments like MATLAB, and all its dependencies are freely available for academic users. Compared to existing tools, PSAMM significantly reduced the running time of constraint-based analysis and enabled flexible settings of simulation parameters using simple one-line commands. The integration of heterogeneous, model-specific annotation information in PSAMM is achieved with a novel format of YAML-based model representation, which has several advantages, such as providing a modular organization of model components and simulation settings, enabling model version tracking, and permitting the integration of multiple simulation problems. PSAMM also includes a number of quality checking procedures to examine stoichiometric balance and to identify blocked reactions. Applying PSAMM to 57 models collected from current literature, we demonstrated how the software can be used for managing and simulating metabolic models. We identified a number of common inconsistencies in existing models and constructed an updated model repository to document the resolution of these inconsistencies.

Fig 1. Comparing the running time of different functions in PSAMM and COBRA.

The running time for both the PSAMM overall function (blue) and the PSAMM problem-solving steps (red) were calculated, and the running time for COBRA only included the problem-solving step (green). Each value represents a median of seven simulation runs using the same specifications, and the error bars indicate the 25th and the 75th percentiles.

Fig 2. Overview of the internal workflow in PSAMM.

The five main components include: (1) user interface, (2) model input/output, (3) model representation, (4) linear programming utilities, and (5) model checking/simulation. Connections among these components form the internal workflow of PSAMM.

Fig 3. An illustration of the PSAMM YAML format.

(a) This diagram shows an example of the YAML model format, which includes a central model definition (model.yaml) and multiple annotation files. Each box indicates a file with a possible filename indicated above the box, and the text within is a snapshot of the file content. (b) An example showing how changes can be tracked in a PSAMM YAML file (biomass.yaml) using the Git version control system in command line. The text highlighted in red indicates the stoichiometry of the compound arg-L in an old version of the biomass function, while the text highlighted in green indicates the updated value in a new version of the model. Additional examples of applying Git version control on the YAML format are provided in the supplemental materials (S2–S5 Texts).

Fig 4. Distribution of blocked reactions in metabolic pathways.

The GEMs were represented in each metabolic pathway as a solid circle. The color of the circles corresponds to the year in which a GEM was published (color legend was shown on the right, and the year of publication ranges from 2003 to 2014). The area of the circle is proportional to the total number of reactions in the pathway, and its vertical position indicates the fraction of reactions that are blocked. The median fractions were indicated by a red mark for each pathway, and models discussed in the main text were highlighted.

Fig 5. A diagram illustrating the modular representation of model components in the YAML format.

The data structure is divided into the static components of model annotation (A) and the dynamic components of simulation settings (B). The reaction and compound annotation databases are associated with a number of required (highlighted in black, e.g. “- id” and “- equation” for reactions) and optional (gray) data entries, and user-defined, model-specific data entries are permitted in the annotation databases. The simulation settings can be represented with various combinations of the model, limits, and media files. Alternative conditions may be defined using a number of alternative modules that can be switched with one another.