Recent technological developments, such as sensors, Next Generation Sequencing (NGS Next Generation Sequencing (Next Generation Sequencing)), and genomics techniques, are providing more and more complex data on the Dutch BIG Law Individual Health care Professions Act (BIG is Dutch acronym for so-called ‘Wet op de beroepen in de individuele gezondheidszorg’). To analyse these large amounts of data, new methods are needed. These include machine learning techniques, methods for static prediction and NGS algorithms. RIVM will investigate which techniques are suitable for RIVM to train itself in them.
The following research projects are launched within the SPR Strategic Programme RIVM (Strategic Programme RIVM) theme "Collections analysis of data".
RIVM National Institute for Public Health and the Environment produces on average two to four native applications for mobile phones each year. To ensure the quality of new apps and the corresponding data, RIVM will develop a basic version of “a generic approach and tooling for developing mobile (native) Apps”: the App Factory.
This approach leads to an improved and more uniform quality of new RIVM apps. It improves the professional look and feel of RIVM apps and contributes to users’ trust in RIVM. It will enable re-use of app functionalities in other RIVM domains, saving development and maintenance costs.
The intended result is a basic version that can be expanded with additional functionalities. These functionalities will be developed and tested in close collaboration with future operators (RIVM staff and partners) and end-users such as citizens or participants in research studies.
Researchers and policymakers can use the basic version of the App Factory for their research projects or during incidents.
RIVM National Institute for Public Health and the Environment is investigating why people living in the vicinity of intensive livestock farming are more likely to develop infections of the lower respiratory tract. Because they are exposed to several substances and micro-organisms at the same time, it is more difficult to determine cause and effect. Therefore, RIVM is developing new methods to determine the health effects when people are exposed to multiple sources at the same time.
Intensive livestock farming is associated with complex environmental challenges, represented by increased air concentrations of particulate matter, of chemicals such as ammonia, and of infectious microbes. Populations living near goat farms have been shown to carry an increased risk of infectious diseases. As inhalation is the major route of exposure in this case, this route will be the focus of the case study in COMPAIR.
The research makes use of existing air measurement networks and research cohorts, such as the Pienter cohorts, and of laboratory research (in vitro models). In addition, new methods are needed to analyse large amounts of complex data (in-silico models). We will develop these new methods using bioinformatics and machine learning, in collaboration with the AMALGAM project.
RIVM National Institute for Public Health and the Environment will develop methods for managing large amounts of data (big data). This research project focuses on the use of machine learning and the analysis of data provided by Next Generation Sequencing (NGS Next Generation Sequencing (Next Generation Sequencing) Next Generation Sequencing ) with the emphasis on data from the microbiome. In this way, AMALGAM supports three other projects of the SPR Strategic Programme RIVM (Strategic Programme RIVM) Strategic Programme RIVM theme "Exposure and Health": COMPAIR, COMPLEXA, and TRIUMPH. AMALGAM is also part of the SPR theme "Collection and analysis of data".
Due to the digitization of society, large data streams are becoming available, which might potentially contain valuable information that can be used by RIVM in the execution of its tasks. In addition, the amount of information generated within RIVM by new technologies in biology, such as Next Generation Sequencing (NGS), is increasing. RIVM needs more knowledge and experience with analysing such data.
First, RIVM will explore which statistical and machine learning methods are available. Second, we will select the most suitable method for our goals, and finally we will test this method by applying it to a data set. Where possible, we choose an application that will also be analysed in COMPAIR, COMPLEXA or TRIUMPH. For the microbiome data, which will only become available at a later stage in these projects, data from RIVM’s VEGA study (study on ESBL-producing bacteria among vegetarians and meat eaters) will be used.