Data in agronomy and other life sciences are often sparse and longitudinal and contain inherent uncertainty that needs to be taken into account. For practical reasons, the results of exploratory analysis of data of this kind should be presented in a way that is interpretable and accessible to scientists in the field. Latent Factor Models can be quite useful to expose the underlying structure of a data set (see West, 2003). A Bayesian framework opens the possibility of incorporating expert knowledge and information about the level of uncertainty in the form of prior distributions (see Rowe, 2000; Minka, 2000; or Ghosh \& Dunson, 2008), and the ability to recover posterior density quantiles. nnOur R package performs Bayesian Latent Factor Analysis on longitudinal data and includes novel graphics to visualize data uncertainty. The Bayesian inference is done using a No-U-Turn-Sampler (see Hoffman \& Gelman, 2011) with the \textit{rstan} package, the R wrapper for the STAN programming language (\url{http://mc-stan.org/}). The package provides graphics that resemble the outputs of classical Principal Component Analysis (a close relative of Latent Factor Models, see Tipping \& Bishop, 1999), but with integrated projection uncertainty regions, enabling facilitated interpretation of the effect of uncertainty on the analysis results.