SBML2HYB: a Python interface for SBML compatible hybrid modeling

Abstract

Summary: Here, we present sbml2hyb, an easy-to-use standalone Python tool that facilitates the conversion of existing mechanistic models of biological systems in Systems Biology Markup Language (SBML) into hybrid semiparametric models that combine mechanistic functions with machine learning (ML). The so-formed hybrid models can be trained and stored back in databases in SBML format. The tool supports a user-friendly export interface with an internal format validator. Two case studies illustrate the use of the sbml2hyb tool. Additionally, we describe HMOD, a new model format designed to support and facilitate hybrid models building. It aggregates the mechanistic model information with the ML information and follows as close as possible the SBML rules. We expect the sbml2hyb tool and HMOD to greatly facilitate the widespread usage of hybrid modeling techniques for biological systems analysis.

Availability and implementation: The Python interface, source code and the example models used for the case studies are accessible at: https://github.com/r-costa/sbml2hyb.

Supplementary information: Supplementary data are available at Bioinformatics online.

Fig. 1.

SBML compatible hybrid modeling pipeline. (A) Overview of the sbml2hyb pipeline. Stored SBML (mechanistic) models in databases are converted to the HMOD format by the sbml2hyb tool. The user inputs the information of the ML component in the sbml2hyb interface (input/output variables and Keras neural network file in H5 format), which is then automatically added to the HMOD file. The resulting hybrid model HMOD file is reconverted to SBML. The hybrid model in SBML format is stored back in databases; (B) Simplified illustration of a hybrid model in SBML format generated by Cytoscape with the cy3sbml app (Konig et al., 2012). On the mechanistic side of the model (left of the image), the larger circles represent the different species of the model, the black squares represent the reactions and the large rectangle refers to the single compartment in this case. On the machine learning side (right of the image), each of the small green circles is a calculation carried out by the ANN, while each blue diamond represents the results of those calculations (hidden and output layers). The small blue circles are the final output of the network, which, in this case, is the value that is assigned to each of the reaction rates