Mixture modeling for joint analysis of survival, discrete, and continuous data

Mixture modeling is commonly used to model categorical latent variables that represent subpopulations in which population membership is unknown but can be inferred from the data. In relatively recent years, the potential of finite mixture models has been applied in time-to-event data. However, the commonly used survival mixture model assumes that the effects of the covariates involved in failure times differ across latent classes, but the covariate distribution is homogeneous. The aim of this dissertation is to develop a method to examine time-to-event data in the presence of unobserved heterogeneity under a framework of mixture modeling. A joint model is developed to incorporate the latent survival trajectory along with the observed information for the joint analysis of a time-to-event variable, its discrete and continuous covariates, and a latent class variable. It is assumed that the effects of covariates on survival times and the distribution of covariates vary across different latent classes. The unobservable survival trajectories are identified through estimating the probability that a subject belongs to a particular class based on observed information. We applied this method to a Hodgkin lymphoma study with long-term follow-up and observed four distinct latent classes in terms of long-term survival and distributions of prognostic factors. Our results from simulation studies and from the Hodgkin lymphoma study demonstrated the superiority of our joint model compared with the conventional survival model. This flexible inference method provides more accurate estimation and accommodates unobservable heterogeneity among individuals while taking involved interactions between covariates into consideration.^

Experimental growth characteristics, cell size, yessotoxin values and bioassay data of Protoceratium reticulatum (Dinophyceae)

Harmful algal blooms are mainly caused by marine dinoflagellates and are known to produce potent toxins that may affect the ecosystem, human activities and health. Such events have increased in frequency and intensity worldwide in the past decades. Numerous processes involved in Global Change are amplified in the Arctic, but little is known about species specific responses of arctic dinoflagellates. The aim of this work was to perform an exhaustive morphological, phylogenetical and toxinological characterization of Greenland Protoceratium reticulatum and, in addition, to test the effect of temperature on growth and production of bioactive secondary metabolites. Seven clonal isolates, the first isolates of P. reticulatum available from arctic waters, were phylogenetically characterized by analysis of the LSU rDNA. Six isolates were further characterized morphologically and were shown to produce both yessotoxins (YTX) and lytic compounds, representing the first report of allelochemical activity in P. reticulatum. As shown for one of the isolates, growth was strongly affected by temperature with a maximum growth rate at 15 °C, a significant but slow growth at 1 °C, and cell death at 25 °C, suggesting an adaptation of P. reticulatum to temperate waters. Temperature had no major effect on total YTX cell quota or lytic activity but both were affected by the growth phase with a significant increase at stationary phase. A comparison of six isolates at a fixed temperature of 10 °C showed high intraspecific variability for all three physiological parameters tested. Growth rate varied from 0.06 to 0.19 per day, and total YTX concentration ranged from 0.3 to 15.0 pg YTX/cell and from 0.5 to 31.0 pg YTX/cell at exponential and stationary phase, respectively. All six isolates performed lytic activity; however, for two isolates lytic activity was only detectable at higher cell densities in stationary phase.