Abstract
BackgroundHIV/AIDS is responsible for the deaths of one million people every year. Although mathematical modeling has provided many insights into the dynamics of HIV infection, there is still a lack of accessible tools for researchers unfamiliar with modeling techniques to apply them to their own clinical data.ResultsHere we present ushr, a free and open-source R package that models the decline of HIV during antiretroviral treatment (ART) using a popular mathematical framework. ushr can be applied to longitudinal data of viral load measurements, and provides processing tools to prepare it for computational analysis. By mathematically fitting the data, important biological parameters can then be estimated, including the lifespans of short and long-lived infected cells, and the time to reach viral suppression below a defined detection threshold. The package also provides visualization and summary tools for fast assessment of model results.Conclusionsushr enables researchers without a strong mathematical or computational background to model the dynamics of HIV using longitudinal clinical data. Increasing accessibility to such methods may facilitate quantitative analysis across a broader range of independent studies, so that greater insights on HIV infection and treatment dynamics may be gained.
Highlights
HIV/AIDS is responsible for the deaths of one million people every year
For the remainder of the paper we focus mainly on the biphasic and single phase models for reverse-transcriptase inhibitor (RTI)/protease inhibitor (PI); further information on implementing the triphasic model for integrase inhibitor (II) may be found in the package documentation
In the case of RTI/PIs, some individuals may have large differences in viral load between the first and second measurements. This is common with sparse clinical data and suggests an unobserved transition from the fast to the slow decay phase. To prevent such occurrences biasing the estimated slope of decay when fitting the single phase model, we remove the first measurement if the difference in viral load is greater than a specified threshold
Summary
HIV/AIDS is responsible for the deaths of one million people every year. mathematical modeling has provided many insights into the dynamics of HIV infection, there is still a lack of accessible tools for researchers unfamiliar with modeling techniques to apply them to their own clinical data. Using the canonical biphasic model of viral decay, the package estimates the lifespans of short and long-lived Model fitting We obtain independent parameter estimates, with 95% confidence intervals, for each subject by fitting either the biphasic, single phase, or triphasic model to the corresponding viral load data using maximum likelihood optimization (as described previously [13]).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
