An algorithm to detect and communicate the differences in computational models describing biological systems

Abstract

Repositories, such as the BioModels Database and the Physiome Model Repository support the reuse of models and ensure transparency about results in publications linked to those models. With thousands of models available, a framework to track the differences between models and their versions is essential to compare and combine models. Difference detection allows users to study the history of models but also helps in the detection of errors and inconsistencies. However, current repositories lack suitable methods to track a model's development over time. Consequently, researchers have problems to grasp the differences between models and their versions.Focusing on SBML and CellML, we developed an algorithm to accurately detect and describe differences between versions of a model with respect to (i) the models' encoding, (ii) the structure of biological networks, and (iii) mathematical expressions. Our method is implemented in a comprehensive and open library called BiVeS. Our work facilitates the reuse and extension of existing models. It also supports collaborative modelling. Finally, it contributes to better reproducibility of modelling results and to the challenge of model provenance.Our algorithm is the first tailor-made detector of differences between versions of computational models in standard formats.

Harmonizing semantic annotations for computational models in biology

“…Semantic annotations can also be used to quantify the similarity between model elements 27,28,[50][51][52][53][54] , helping to ensure that the user is presented with the most relevant models following a search. Other types of annotations, such as those that capture provenance information [55][56][57][58] , can also help researchers determine which models best meet their research needs and make comparisons between different model versions 59 .…”

Notions of similarity for computational biology models

“…If models are encoded as computer programs, e. g. in MATLAB, a comparison on the syntactic level can be performed using tools for difference detection in software code (e.g., diff). Due to the lack of a predetermined structure of general Several algorithms use this reduced approach of determining the similarity of models (or model versions) by comparing their syntactic structure, e. g. [19,23,24]. Some software tools compare representations of a model while taking the specific syntactic structure and its corresponding biological meaning into account, while others compare the representations directly and subsequently interpret the results with respect to the biology.…”

“…Some software tools compare representations of a model while taking the specific syntactic structure and its corresponding biological meaning into account, while others compare the representations directly and subsequently interpret the results with respect to the biology. For example, XML patches [23] are generated to compare models at the XML level only, while the BiVeS tool described in [19] also considers the structure of the actual model representation format.…”

“…The comparison of model versions can help to keep track of the model's evolution in time, and it identifies points at which a model underwent major changes[42]. BiVeS[19] is a software library that aligns the XML encodings of two model versions, identifies and interprets changes, and measures the changes' impact. BiVeS also considers characteristics of the model encoding format and differentiates between SBML and CellML.…”

MAMMOTh: a new database for curated MAthematical Models of bioMOlecular sysTems

“…However, reuse, expansion, or modification of the developed models is a labor-intensive process (29)(30)(31)(32). Usually the models are designed as complete integrated systems and there are no tools available for researchers to modify their structural and/or functional contents in an automatic or semiautomatic way (31). The decomposition of the models into their elementary subsystems is an alternative approach to solve this problem.…”