To develop computer programs to put Framingham Study data into a format that can be read into the demographic programs; to revise the analysis programs to increase the speed with which they process the data; and to develop a mechanized approach to scanning the results and identifying SNPs that appear to be associated with longevity.
The search for genes associated with longevity is moving from the analysis of data for a small number of candidate genes to analysis of data from very large genome scans. Recently the Framingham Study has made available to researchers a data set covering about 10,000 individuals with a scan of 550,000 markers. The standard approach used by genetic epidemiologists to analyzing this type of data is to apply Cox regressions to the data on survival for a period of observation following the collection of the genetic material. A much more efficient method relies on a demographic model that can also make use of the data on genotype by age at the time the genetic material was collected. This approach has recently been applied to data for 1700 individuals with approximately 700 genetic markers. This application seeks support for an extension of the existing demographic model to make use of data on the survival status of the parents of genotyped individuals who were not themselves genotyped. We also seek support to develop more efficient methods to running the analysis which could then be applied to the much larger data set that is now available.