Surveillance and claims data
- Puerto Rico fungal-spore and respiratory-virus surveillance
- Puerto Rico Department of Health / ASES claims and pharmacy data
- Meteorology and wearable exposure streams
Research
The scientific approaches we use to accomplish our vision and mission.
Exposure science
As stated by the US National Academies of Science in Exposure Science in the 21st Century: A Vision and a Strategy, exposure science links (a) the origin of a pollutant and its concentration from a given source, (b) human behavior during or as a result of the exposure, (c) the interaction between the pollutant and the human subject, and (d) the outcome of that interaction. Depending on the question and the project — often collaborative — we address two or more of these components.
Human-based immunotoxicology
To align biomarker profiles with environmental pollutant exposure, we employ human-based immunological approaches. Our principal investigator has a long track record of expanding the utility of immunological assays into new areas of research, recognized among others by the Lush Prize Young Researcher award.
Computational data science and data management
It is a difficult case for science when there are relevant findings but the statistical approaches are not reproducible. At RIPLRT we use R, Python, and MATLAB to keep our data science reproducible. We publish analysis scripts, and often raw data, in open repositories including GitHub, the Open Science Framework, FigShare, and Zenodo.
The methods have evolved from pipettes to pipelines. RIPLRT began in 2018 as a wet-lab immunology group; since 2021 its center of gravity has shifted toward a hybrid computational and translational model that pairs molecular work with repository-based discovery and population-level forecasting. The mission — understanding environmental drivers of respiratory and immune health and translating that understanding into equity — has been the constant since day one.
Where the science lives
Data and methods
Our near-term implementation cycle emphasizes real-world and computational validation using existing surveillance, healthcare, and biomarker-rich repositories alongside new laboratory work — all within a reproducibility and equity-stratified analysis framework.
National repositories
- BioLINCC and BioData Catalyst
- All of Us
- MESA, Framingham, and AsthmaNet
- NHANES
Analytic methods
- AI/ML with SHAP explainability
- Distributed-lag and mixed-effects models
- GWAS and polygenic risk scores
- Causal mediation analysis
- Multi-agent AI architectures
- Equity-aware performance metrics
Full detail on our scientific identity, data resources, and reproducibility standards is in the RIPLRT Institute Research Group Manual.
Active themes
Where our projects are focused
Post-disaster indoor environments
Indoor fungal amplification and pro-inflammatory potential of settled dust in homes water-damaged during Hurricane María in Puerto Rico.
Aeroallergens and respiratory outcomes
Ambient fungal spores and pollen as predictors of asthma healthcare utilization and respiratory virus incidence, using time-series and machine-learning models.
Microbiome and respiratory health
Nasal and salivary microbiome profiling in infants and adults, including non-invasive collection methods developed with our collaborators.