Skip to main content
. 2008 Oct 21;29(7):823–864. doi: 10.1210/er.2008-0005

Table 2.

Example of ill-posed deconvolution problem

Infused LH profile1 Simulated LH profile2 Sheep LH profile3
A. Starting Estimates
Burst sd (min)a 1 10 30 1 10 30 1 10 30
Half-life (min) 15 50 150 15 50 150 15 50 150
Basal secretionb 0 0.01 0.1 0 0.01 0.1 0 0.01 0.1
Starting concentration 1 1 1 1 1 1 1 1 1
B. Deconvolution results
Frequency (per session) 6 6 zero 16 17 4 zero zero zero
Burst sd (min)a 2.6 2.5 −9.9 0.74 1.3 0.5 −0.58 −100 1.2
Half-life (min) 61 49 551 53 50 334 2.2 2.3 391
Basal secretionb 0.03 0.05 0.009 0.02 0.02 0.003 3.0a 2.8 0.02
Mass/burstb 5.9 5.5 zero 3.0 2.9 1.7 zero zero zero

Analyses used a web-based automated deconvolution (AutoDecon) program downloaded from http://mljohnson.pharm.virginia.edu/downloads.html on 11/07/07. The three 10-min LH profiles are listed in the Appendix (published as supplemental data on The Endocrine Society’s Journals Online web site at http://mend.endojournals.org) as ″Lhinfused,″ ″Lhsimulated,″ and ″Lhsheep.″ 

1, 2, 3 True pulse numbers were 6, 16, and 7, respectively, based upon specified LH injections, explicit simulations, and portal-venous GnRH pulses. 

a

sd of Gaussian burst. 

b

Concentration units per minute for basal secretion and concentration units for mass.