processes
Review
The Spectrum of Mechanism-Oriented Models and
Methods for Explanations of Biological Phenomena
C. Anthony Hunt 1, *, Ahmet Erdemir 2 , William W. Lytton 3 , Feilim Mac Gabhann 4 ,
Edward A. Sander 5 , Mark K. Transtrum 6 and Lealem Mulugeta 7
1
Department of Bioengineering and Therapeutic Sciences, University of California, San Francisco, CA 94143, USA
2
Department of Biomedical Engineering and Computational Biomodeling Core, Lerner Research Institute,
Cleveland Clinic, Cleveland, OH 44195, USA; erdemira@ccf.org
3
Departments of Neurology and Physiology and Pharmacology, SUNY Downstate Medical Center,
Department Neurology, Kings County Hospital Center, Brooklyn, NY 11203, USA;
billl@neurosim.downstate.edu
4
Institute for Computational Medicine and Department of Biomedical Engineering, Johns Hopkins
University, Baltimore, MD 21218, USA; feilim@jhu.edu
5
Department of Biomedical Engineering, University of Iowa, Iowa City, IA 52242, USA;
edward-sander@uiowa.edu
6
Department of Physics and Astronomy, Brigham Young University, Provo, UT 84602, USA;
mktranstrum@byu.edu
7
InSilico Labs LLC, Houston, TX 77002, USA; lealem.mulugeta@gmail.com
* Correspondence: a.hunt@ucsf.edu; Tel.: +1-415-476-2455
Received: 15 April 2018; Accepted: 6 May 2018; Published: 14 May 2018
Abstract: Developing and improving mechanism-oriented computational models to better explain
biological phenomena is a dynamic and expanding frontier. As the complexity of targeted
phenomena has increased, so too has the diversity in methods and terminologies, often at the
expense of clarity, which can make reproduction challenging, even problematic. To encourage
improved semantic and methodological clarity, we describe the spectrum of Mechanism-oriented
Models being used to develop explanations of biological phenomena. We cluster explanations
of phenomena into three broad groups. We then expand them into seven workflow-related
model types having distinguishable features. We name each type and illustrate with examples
drawn from the literature. These model types may contribute to the foundation of an ontology
of mechanism-based biomedical simulation research. We show that the different model types
manifest and exert their scientific usefulness by enhancing and extending different forms and degrees
of explanation. The process starts with knowledge about the phenomenon and continues with
explanatory and mathematical descriptions. Those descriptions are transformed into software
and used to perform experimental explorations by running and examining simulation output.
The credibility of inferences is thus linked to having easy access to the scientific and technical
provenance from each workflow stage.
Keywords: computational model; explanatory model; hybrid model; mechanism; mechanistic model;
modeling methods; provenance; workflow; systems modeling; simulation
1. Introduction
Within the large context of biological systems modeling and analysis, developing and
improving mechanism-oriented computational models to better explain complex biological phenomena
are expanding. As the complexity of the phenomena to be explained has increased, the diversity
in methods and terminologies has also increased, often at the expense of clarity, which can
Processes 2018, 6, 56; doi:10.3390/pr6050056
www.mdpi.com/journal/processes
181
Review
The Spectrum of Mechanism-Oriented Models and
Methods for Explanations of Biological Phenomena
C. Anthony Hunt 1, *, Ahmet Erdemir 2 , William W. Lytton 3 , Feilim Mac Gabhann 4 ,
Edward A. Sander 5 , Mark K. Transtrum 6 and Lealem Mulugeta 7
1
Department of Bioengineering and Therapeutic Sciences, University of California, San Francisco, CA 94143, USA
2
Department of Biomedical Engineering and Computational Biomodeling Core, Lerner Research Institute,
Cleveland Clinic, Cleveland, OH 44195, USA; erdemira@ccf.org
3
Departments of Neurology and Physiology and Pharmacology, SUNY Downstate Medical Center,
Department Neurology, Kings County Hospital Center, Brooklyn, NY 11203, USA;
billl@neurosim.downstate.edu
4
Institute for Computational Medicine and Department of Biomedical Engineering, Johns Hopkins
University, Baltimore, MD 21218, USA; feilim@jhu.edu
5
Department of Biomedical Engineering, University of Iowa, Iowa City, IA 52242, USA;
edward-sander@uiowa.edu
6
Department of Physics and Astronomy, Brigham Young University, Provo, UT 84602, USA;
mktranstrum@byu.edu
7
InSilico Labs LLC, Houston, TX 77002, USA; lealem.mulugeta@gmail.com
* Correspondence: a.hunt@ucsf.edu; Tel.: +1-415-476-2455
Received: 15 April 2018; Accepted: 6 May 2018; Published: 14 May 2018
Abstract: Developing and improving mechanism-oriented computational models to better explain
biological phenomena is a dynamic and expanding frontier. As the complexity of targeted
phenomena has increased, so too has the diversity in methods and terminologies, often at the
expense of clarity, which can make reproduction challenging, even problematic. To encourage
improved semantic and methodological clarity, we describe the spectrum of Mechanism-oriented
Models being used to develop explanations of biological phenomena. We cluster explanations
of phenomena into three broad groups. We then expand them into seven workflow-related
model types having distinguishable features. We name each type and illustrate with examples
drawn from the literature. These model types may contribute to the foundation of an ontology
of mechanism-based biomedical simulation research. We show that the different model types
manifest and exert their scientific usefulness by enhancing and extending different forms and degrees
of explanation. The process starts with knowledge about the phenomenon and continues with
explanatory and mathematical descriptions. Those descriptions are transformed into software
and used to perform experimental explorations by running and examining simulation output.
The credibility of inferences is thus linked to having easy access to the scientific and technical
provenance from each workflow stage.
Keywords: computational model; explanatory model; hybrid model; mechanism; mechanistic model;
modeling methods; provenance; workflow; systems modeling; simulation
1. Introduction
Within the large context of biological systems modeling and analysis, developing and
improving mechanism-oriented computational models to better explain complex biological phenomena
are expanding. As the complexity of the phenomena to be explained has increased, the diversity
in methods and terminologies has also increased, often at the expense of clarity, which can
Processes 2018, 6, 56; doi:10.3390/pr6050056
www.mdpi.com/journal/processes
181