7 replies to this topic

[harishkk]

Hello All,

Can any one suggest me where the error is while iam executing the NLME model built using DME,

Iam attaching the PHX Prj file for your reference

 

Regards

Harish

Attached Files

•  Error while executing.phxproj   611.46KB   442 downloads

[Simon Davis]

Hi Harish, I opened up your project (I see it was created in version 6.4) and saw in the warnings file;

 

Error in Population:

'".\NLME7.exe"' is not recognized as an internal or external command,

operable program or batch file.

 

Model execution failed or cancelled

 

I tried executing as it stood and got the same message.  I then looked a little deeper and noted acoupel of things;

 

1) you had not submitted your initial estimates for  Fixed Effects (remember you must click the little arrows when using this visual tool)

2) Also you had entered an AMT on every time, whereas I suppose your dose was really only given once at time zero

>so I removed the other doses.

 

  However it still did not run, so I made a copy of the model and just to check how it was set up I reverted one step to built in model and it ran fine.

  Please take a look at the copy and then if you need to make some changes copy that object first and them make changes to the daughter so you still have something 'good' to go back to.

 

 Simon.

Attached Files

•  Harish.phxproj   1.25MB   448 downloads

[smouksassi1]

you had two compartments with the same name A0.

When you fix this it will work.

currently the engine will run but the model need tweaking to correct some of the evident bias seen.

attached a projects that I quickly edited

 

Regards,

 

Samer

Attached Files

•  Error while executing.phxproj   1.48MB   444 downloads

[harishkk]

Hello Simon and Sameer

 

Ya I have realised these yesterday itself, agreed with your observations

 

Thanks

Harish 

[serge guzy]

Dear Harish,

I looked at your project and I see some modeling issues. I corrected the model to make it more attractive

1: Look at the model building approach. I started with NCA, then 1c, then 2 c then 2c with tlag and after that I saw that in fact we have 2 peaks with 2 different tlags. I added the 2 tlags and it looks better.

 

Because of the small number of patients and data, it may be problematic to conclude statistically we have 2 peaks but if you look at each individual, there is really a sense of multiple absorption.

I use also multiplicative error which is better than additive for PK data.

The project is attached.

 I strongly suggest you to go to one of our population PK/PD course using NLME to learn about the optimal model building strategies and complex modeling.

 

https://www.certara.com/training

best Regards

Serge

Attached Files

•  harish_serge_corrected.phxproj   4.19MB   451 downloads

Edited by Simon Davis, 09 October 2015 - 10:21 AM.
remove unnecessary quoting, add link to training page...

[harishkk]

Dear Sergy

Thanks for the expert's eye and giving a new dimension to my data. Regarding NLME training, Iam trying to use PHX NLME for my work off late and i have attended few workshop's by Joga at India. It will take a bit time for me to get to terms with PHX NLME

But until then i require some of your help 

 

Thanks & Regards

Harish

[serge guzy]

Dear Sergy

Thanks for the expert's eye and giving a new dimension to my data. Regarding NLME training, Iam trying to use PHX NLME for my work off late and i have attended few workshop's by Joga at India. It will take a bit time for me to get to terms with PHX NLME

But until then i require some of your help 

 

Thanks & Regards

Harish

My pleasure. note that I am in India right now in Mumbai giving a NLME workshop tomorrow.

You better go to an advanced workshop we will provide in 2016.

Best Regards

Serge

[harishkk]

Dear Sergy,

Thanks for your open support, I know that you are here in India giving a workshop course on Modelling at Mumbai, as some of my friends are attending

 

If there are any plans by Certara to plan any advanced course in India, it would be helpful, hope it would be done

 

Regards

Harish