ABSTRACT
Domain modelling languages (DMLs) grow and change over time. These languages are artefacts that are developed within communities via multiple participants. Methods, associated with the emerging DMLs, also need to be supported and need adaptation, informed by practice. This study refers to a DML called DEMO (Design and Engineering Methodology for Organizations), of which the language specification evolved from the DEMO Specification Language (DEMOSL) version 3 to version 4. We adapt a method, called the story-card-method (SCM), to accommodate DEMOSL 4. Also, the previous DEMOSL 3-based SCM implied physical interaction between participants, using sticky notes to create a shared understanding, whereas the adapted SCM has to facilitate a digital way-of-collaboration due to COVID-19 restrictions. We re-visit participant feedback from the initial version of the SCM and demonstrate how we applied design science research to design an adapted SCM as the main contribution of this article. In addition, we evaluate whether the adapted SCM is useful in providing ample guidance in compiling a Coordination Structure Diagram (CSD) in a collaborative way, and we evaluate the quality of the CSDs. Finally, we demonstrate how the CSD can be used within a low-code-development ecosystem to structure user stories. © 2022, Springer Nature Switzerland AG.
ABSTRACT
One of our greatest present challenges are designing vaccines against SARS COV2 and its variants. Rational vaccine design uses computational methods prior to development of a vaccine for testing in animals and humans the latest methods in rational vaccine design use machine learning techniques to predict binding affinity and antigenicity but offer the researchers only isolated stand-Alone tools. A difficulty that software engineers and data scientist face in development of tools for doctors and researchers is their lack of knowledge of the medical domain. This paper presents a set of domain model developed in collaboration between software engineers and a medical researcher in the process of building a tool scientists could use to predict binding affinity and antigenicity of potential designs of SARS COV2 vaccines. A domain model visualizes the real-world entities and their interrelationships, that together define the domain space. This domain model will be useful to other software engineers trying to predict other characteristics of vaccines, such as potential autoimmunity response. © 2021 IEEE.