Mathematical modeling of epidemiology of Alzheimer's disease.

 
   Lead Investigator:    Natalia Komarova
   Institution      :    University of California Irvine
   E-Mail           :    komarova@uci.edu
   Proposal ID      :    312



   Proposal Description:
   Aim 1: Create a number of models that describe the development of AD, as a series of m steps (with varying m, and varying rates of change). These models are stochastic multi-stage processes much like the ones used in the field of oncology [3,2].   Aim 2: Different models of disease initiation and progression will be consistent with different age-incidence curves characteristic of the disease.  We will apply the models to the data and use a model selection procedure to weed out models that do not correspond to reality. Use iterations to select for the best fitting models.   Aim 3: Repeat Aim 2 with different groups of patients. For example, consider splitting the patient population by demographic factors (male/female, race), by family histories, by genetic status (if possible). Even more interestingly, by examining the longitudinal progression data, split the patents in the groups of fast and slow progressors (as we did in [5,6]) and apply the analysis of Aims 1-2 to these groups. Do fast and slow progressors seem to have different underlying processes governing disease onset and development?