Re: [NMusers] Validation and power of the Mixture Model

From: L. Chakradhar <chakradhar_shetty_at_yahoo.co.in>
Date: Thu, 15 Oct 2015 12:17:54 +0000 (UTC)

Dear Hui,
The below article will be of interest to you:
Mixture models and subpopulation classification: a PK simulation study and =
application to metoprolol CYP2D6 phenotype. J Pharmacokinet Pharmacody=
n. 2007 Apr;34(2):141-56.
RegardsChakri



 


     On Wednesday, 14 October 2015 1:34 PM, "HUI, Ka Ho" <matthew.hui_at_link.=
cuhk.edu.hk> wrote:
   

  <!--#yiv3867282658 _filtered #yiv3867282658 {font-family:PMingLiU;panose-=
1:2 2 5 0 0 0 0 0 0 0;} _filtered #yiv3867282658 {font-family:"Cambria Math=
";panose-1:2 4 5 3 5 4 6 3 2 4;} _filtered #yiv3867282658 {font-family:Cali=
bri;panose-1:2 15 5 2 2 2 4 3 2 4;} _filtered #yiv3867282658 {font-family:P=
MingLiU;panose-1:2 2 5 0 0 0 0 0 0 0;}#yiv3867282658 #yiv3867282658 p.yiv38=
67282658MsoNormal, #yiv3867282658 li.yiv3867282658MsoNormal, #yiv3867282658=
 div.yiv3867282658MsoNormal {margin:0cm;margin-bottom:.0001pt;font-size:12.=
0pt;font-family:"Calibri", sans-serif;}#yiv3867282658 a:link, #yiv386728265=
8 span.yiv3867282658MsoHyperlink {color:#0563C1;text-decoration:underline;}=
#yiv3867282658 a:visited, #yiv3867282658 span.yiv3867282658MsoHyperlinkFoll=
owed {color:#954F72;text-decoration:underline;}#yiv3867282658 span.yiv38672=
82658EmailStyle17 {font-family:"Calibri", sans-serif;color:windowtext;}#yiv=
3867282658 .yiv3867282658MsoChpDefault {font-family:"Calibri", sans-serif;}=
#yiv3867282658 _filtered #yiv3867282658 {margin:72.0pt 90.0pt 72.0pt 90.0pt=
;}#yiv3867282658 div.yiv3867282658WordSection1 {}-->Dear all,   I am =
a fresh research student. Previously, I had the experience in developing a =
pharmacogenetics-based population PK model using NONMEM, but this is the fi=
rst time I work with the Mixture Model in NONMEM and I have some questions =
about it.   I have two datasets on hand, where patients’ clea=
rance are believed to be affected by their genotypes. While I have their ge=
notypic information, I also wish to know if the Mixture Model in NONMEM can=
 help me accurately categorize the population even if the genotypic informa=
tion are “hidden” from NONMEM. The results turned out to be=
 unsatisfactory somehow. For the subpopulations with a distinctively differ=
ent typical values of clearance, the sensitivity and specificity can approa=
ch 100%, but for those with less differences, the average accuracy drops to=
 60-70%. Although it is not difficult to understand that the computer will =
not be able to categorize these subjects when they have similar parameters =
(either mean values too close or variances too large…), I am wonder=
ing if there is any general approach to utilize the best out of the Mixture=
 Model function.   Regarding the power of the Mixture Model, I wonder=
 if there has been any validation done before for datasets with different c=
haracteristics. For examples, is there any previous study that looked into =
the accuracy of the Mixture Model function and can somehow express the typi=
cal accuracy in terms of the difference in, say, the mean plasma levels, be=
tween 2 subpopulations.   Last but not least, it would be great if an=
yone can kindly advise me any good teaching materials about the Mixture Mod=
el in NONMEM.   Sincerely, Matthew Hui

  
Received on Thu Oct 15 2015 - 08:17:54 EDT

This archive was generated by hypermail 2.3.0 : Fri Sep 27 2019 - 16:46:38 EDT