Biology Reference
In-Depth Information
Secretory cell
Blood vessel
Receptor
Target cell
Hormone
FIGURE 9-1.
Simple model of endocrine function. The secretory cell in the endocrine gland produces the
hormone that is carried by the bloodstream. The hormone binds to a receptor on or in the target
cell. Hormone binding causes a change in the behavior of the target cell.
designing experiments, collecting data, and analyzing hormone secretion
patterns.
So far, our focus has been primarily on model development and data
analyses, largely ignoring the data-collection process. For two reasons,
however, data collection is not straightforward when determining
hormone secretion patterns. First, it is generally not possible to collect
data directly from the endocrine glands, where the hormones are
secreted, for reasons discussed below. Instead, information about
hormone secretion is inferred from data representing the hormone
concentration in the blood. Second, even these data do not directly give
an accurate picture of the secretion patterns, because once the hormone
is secreted and has entered the bloodstream, its physiological
elimination from the blood (because of binding, excretion, and/or
biotransformation) begins immediately. Figuratively speaking, such data
only provide a glimpse through a ''dirty window,'' as the hormone
secretion patterns are distorted because of the ongoing elimination
processes. We need to ''clean'' the window by removing the effect of
hormone elimination in order to be able to ''see'' the actual secretion
amplitudes and frequencies.
We shall present a number of mathematical methods, both classical and
novel, aimed at quantifying various aspects of this problem. The
methods are divided into two groups. The first covers statistical
approaches for analysis of hormone concentrations as observed in the
blood. The second employs deconvolution methods to deduce hormone
secretion patterns from the hormone concentration in the blood.
Although some of the classical methods, such as the Fourier methods
outlined in this chapter, are now known to be of limited use for
analyzing general hormone data, we have included them because they
are still routinely used in the literature to analyze specific aspects of
hormone pulsatility.
Several of the mathematical methods we describe require data in the
form of a time series (i.e., measurements of hormone concentration
