|
Activity Number:
|
378
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #307134 |
|
Title:
|
A Bayesian Approach to Modeling Associations between Pulsatile Hormones
|
|
Author(s):
|
Nichole Carlson*+ and Timothy D. Johnson and Morton B. Brown
|
|
Companies:
|
Oregon Health & Science University and University of Michigan and University of Michigan
|
|
Address:
|
Division of Biostatistics, 3181 SW Sam Jackson Park Road, Portland, OR, 97239,
|
|
Keywords:
|
hormone data ; biological modeling ; birth-death MCMC ; pulse association
|
|
Abstract:
|
Many hormones are secreted in pulses. The pulsatile relationship between hormones regulates many biological processes. To understand endocrine system regulation, time series of hormone concentrations are collected. The goal is to characterize pulsatile patterns and associations between the hormones. When the signal-to-noise ratio is large, pulse detection and parameter estimation is difficult with existing approaches. We present a bivariate deconvolution model of pulsatile hormone data using a Bayesian approach, which addresses these issues. We describe a model for a one-to-one, driver-response case and show how birth-death MCMC can be used for estimation. We exhibit that using known pulsatile associations and bivariate fitting improves estimation of the pulse locations and the parameters for each hormone. An example is presented using luteinizing and follicle stimulating hormones.
|
