Luís Bastião Silva1, Alina Trifan2, José Luís Oliveira3. 1. BMD Software, Aveiro, Portugal. Electronic address: bastiao@bmd-software.com. 2. University of Aveiro, DETI/IEETA, Portugal. Electronic address: alina.trifan@ua.pt. 3. University of Aveiro, DETI/IEETA, Portugal. Electronic address: jlo@ua.pt.
Abstract
BACKGROUND AND OBJECTIVE: Data catalogues are a common form of capturing and presenting information about a specific kind of entity (e.g. products, services, professionals, datasets, etc.). However, the construction of a web-based catalogue for a particular scenario normally implies the development of a specific and dedicated solution. In this paper, we present MONTRA, a rapid-application development framework designed to facilitate the integration and discovery of heterogeneous objects, which may be characterized by distinct data structures. METHODS: MONTRA was developed following a plugin-based architecture to allow dynamic composition of services over represented datasets. The core of MONTRA's functionalities resides in a flexible data skeleton used to characterize data entities, and from which a fully-fledged web data catalogue is automatically generated, ensuring access control and data privacy. RESULTS: MONTRA is being successfully used by several European projects to collect and manage biomedical databases. In this paper, we describe three of these applications scenarios. CONCLUSIONS: This work was motivated by the plethora of geographically scattered biomedical repositories, and by the role they can play altogether for the understanding of diseases and of the real-world effectiveness of treatments. Using metadata to expose datasets' characteristics, MONTRA greatly simplifies the task of building data catalogues. The source code is publicly available at https://github.com/bioinformatics-ua/montra.
BACKGROUND AND OBJECTIVE: Data catalogues are a common form of capturing and presenting information about a specific kind of entity (e.g. products, services, professionals, datasets, etc.). However, the construction of a web-based catalogue for a particular scenario normally implies the development of a specific and dedicated solution. In this paper, we present MONTRA, a rapid-application development framework designed to facilitate the integration and discovery of heterogeneous objects, which may be characterized by distinct data structures. METHODS: MONTRA was developed following a plugin-based architecture to allow dynamic composition of services over represented datasets. The core of MONTRA's functionalities resides in a flexible data skeleton used to characterize data entities, and from which a fully-fledged web data catalogue is automatically generated, ensuring access control and data privacy. RESULTS: MONTRA is being successfully used by several European projects to collect and manage biomedical databases. In this paper, we describe three of these applications scenarios. CONCLUSIONS: This work was motivated by the plethora of geographically scattered biomedical repositories, and by the role they can play altogether for the understanding of diseases and of the real-world effectiveness of treatments. Using metadata to expose datasets' characteristics, MONTRA greatly simplifies the task of building data catalogues. The source code is publicly available at https://github.com/bioinformatics-ua/montra.
Authors: João Rafael Almeida; Rosa Gini; Giuseppe Roberto; Peter Rijnbeek; José Luís Oliveira Journal: BMC Med Inform Decis Mak Date: 2019-07-02 Impact factor: 2.796