Rice University doctoral student Jose Palacio is developing an R code package for data analysis and statistical computing that is especially relevant in measuring for variability in wastewater dynamics and infectious disease targeting and trending.
The project is part of ongoing research by statistician Katherine Ensor’s research group at Rice, which involves multi-institutional efforts to translate engineering innovation in wastewater-based epidemiology to guard human health.
Wastewater surveillance has proven to be crucial for monitoring contaminants and tracking viral spread in communities. This study uses a new multivariate state-space model to analyze SARS-CoV-2 RNA in wastewater from 32 treatment plants in Houston that serve 2.2 million people.
The study data analysis includes weekly composite influent samples collected from July 6, 2020, to October 28, 2024, with RNA concentrations log-transformed and missing values handled per standard protocols.
“When considering the size and scale of Houston’s wastewater system, there is variability in finding and tracking viral transmissibility as it relates to wastewater treatment plant flow and evolutionary work processes from wastewater sampling to laboratory and data analysis,” explained Palacio, who is a third-year statistics doctoral student with Ensor as his advisor.
Built for use by Houston Wastewater Epidemiology, the model applies recent technology from the EpiSewer R package for effective reproduction (Rt) trends while integrating step-by-step variability analysis of population wastewater infection dynamics for covid, influenza, and respiratory syncytial virus (RSV).
“Using a multivariate state-space model, we capture wastewater dynamics and viral trends over time. The approach improves interpretability in wastewater-based epidemiology by distinguishing true viral signals from observational noise,” said Palacio.
The analysis includes multivariate autoregressive state-space modeling (MARSS) R package while applying Kalman filtering for estimation. The project is part of developing efforts to scale wastewater-based epidemiology to serve as an early warning system for infection prevention and control for up to 29 potential pathogens.
Palacio’s and Ensor’s work in this area is particularly important as it provides flexibility toward pathogen detection and changes in sampling and laboratory processes as science in wastewater-based epidemiology continues to evolve rapidly.
Ensor, Rice’s Noah G. Harding Professor of Statistics, is known globally for her work in mathematical statistics and methods in computational analysis, specifically in areas of time series and spatial processes, has worked jointly with the City of Houston and the Houston Health Department for more than a decade on projects involving state-space modeling and statistical process control frameworks to improve the environment and human health.
Palacio is a graduate student fellow of Rice’s Ken Kennedy Institute, which supports research related to high-performance computing, computational science & engineering, AI, data science, and machine learning. Palacio graduated summa cum laude with a B.S. in mathematics and a concentration in statistics from the University of Texas Rio Grande Valley in 2020.
Houston Wastewater Epidemiology, a CDC National Wastewater Surveillance System (NWSS) Center of Excellence, is a collaboration involving interdisciplinary teams of engineers, municipal and human health professionals with the Houston Health Department, Rice University, and Houston Public Works.
Shawn Hutchins, Communications Specialist