Re: Block SIZE with SCM

From: Jakob Ribbing Date: December 09, 2019 technical Source: mail-archive.com
Dear Emeric, With the PsN version I have used previously, I can confirm it has worked fine to set items in $SIZES and run scm from that model. I have for example used this: ;; Initial comments to describe the model $SIZES PD=121 MAXIDS=18000 $PROBLEM Description of model $INPUT ... Maybe there was some strange character in your control stream that made the line break disappear? For example, looking down into your previous e-mail I now notice that the dollar($) sign in front of PROBLEM seems strange (my bold, just below): “Here is the beginning of my code : $SIZES LVR=150 PD=-100 SPROBLEM empty model $DATA..." What you put into your original e-mail looks like an S (did not notice that before). Whatever character that is, it seems to be accepted by NONMEM as a dollar sign, but not by PsN (and neither by the different means I have to view your e-mail). If you change that to a proper dollar sign, I think the original setup with PD=-100 will run in scm, as intended? Best wishes Jakob
Quoted reply history
> On 9 Dec 2019, at 09:04, Emeric Sibieude <[email protected]> > wrote: > > Dear Jakob, > > Thank you for your answer. According to PsN documentation PD = -100 means at > least 100 covariates. I tried with the exact number of covariates (PD = 105) > and I have the same issue. > > Yes indeed, I can do it with the default value of PD = 50 because less than > 50 are added but I was wondering if there exists a way if more than 50 > covariates are added during the forward path. > > Best Regards, > > Emeric Sibieude > > Student trainee in Master's degree > > Biopharma | R&D | Global Development | Translational Medicine > > Merck > > Lausanne , Switzerland > > [email protected] <mailto:[email protected]> > > From: Jakob Ribbing <[email protected] > <mailto:[email protected]>> > Sent: jeudi 5 décembre 2019 09:11 > To: Emeric Sibieude <[email protected] > <mailto:[email protected]>>; [email protected] > <mailto:[email protected]> > Subject: Re: [NMusers] Block SIZE with SCM > > Dear Emeric, > > The code has negative PD, maybe PsN has problems interpreting this? > I am not sure how PD=-100 would be interpreted by NONMEM, but try PD=100, > instead. > I think this is what you intended to do? > > As a side note, PsN (mainly to handle old nonmem versions) has functionality > for dropping (=DROP) covariates on $INPUT, if they are not tested in the > current model during SCM - but only among the ones listed in the > configuration as covariates. > Even if you test 50 covariates, it is not often more than ten different will > be included during forward inclusion, so possibly you can do with the default > PD=50? > > Best wishes > > Jakob > > > On 5 Dec 2019, at 08:39, Emeric Sibieude <[email protected] > <mailto:[email protected]>> wrote: > > Dear NMUsers, > > I am performing scm with PsN. My data have more than 50 covariates, this > should be processed by $SIZES PD=100 at the beginning of the NONMEM file. > > However, it seems that scm command modifies this file by putting the block > $PROBLEM at the first line. > > > > Here is the beginning of my code : > > $SIZES LVR=150 PD=-100 > > SPROBLEM empty model > > $DATA ID TIME AMT EVID DV CAT1 CAT2 ... > > > > Here is the code transformed in the LST file > > $SIZES LVR=150 PD=-100 $PROBLEM empty model > > $DATA ID TIME AMT EVID DV CAT1 CAT2 ... > > > > And at the end I have the following error > > NM-TRAN MESSAGES > > AN ERROR WAS FOUND IN THE CONTROL STATEMENTS. > > AN ERROR WAS FOUND ON LINE 1 AT THE APPROXIMATE POSITION NOTED: > $SIZES LVR=150 PD=-100 $PROBLEM empty mod > X > THE CHARACTERS IN ERROR ARE: $ > 20 UNKNOWN OPTION. > > > > Note that this is an issue for estimating the base model only. During the > forward path scm automatically drops the unnecessary covariates keeping only > the ones added at previous steps and the tested one. First estimate the base > model and use the option -base_ofv seems to work however I will have the same > issue if more than 50 covariates are added during the forward path. > > Thank you in advance for any help or any suggestions. > > Best Regards, > > Emeric Sibieude > > Student trainee in Master's degree > > Biopharma | R&D | Global Development | Translational Medicine > > Merck > > Lausanne , Switzerland > > [email protected] <mailto:[email protected]> > > > > > > This message and any attachment are confidential and may be privileged or > otherwise protected from disclosure. If you are not the intended recipient, > you must not copy this message or attachment or disclose the contents to any > other person. If you have received this transmission in error, please notify > the sender immediately and delete the message and any attachment from your > system. Merck KGaA, Darmstadt, Germany and any of its subsidiaries do not > accept liability for any omissions or errors in this message which may arise > as a result of E-Mail-transmission or for damages resulting from any > unauthorized changes of the content of this message and any attachment > thereto. Merck KGaA, Darmstadt, Germany and any of its subsidiaries do not > guarantee that this message is free of viruses and does not accept liability > for any damages caused by any virus transmitted therewith. > > Click http://www.merckgroup.com/disclaimer > http://www.merckgroup.com/disclaimer to access the German, French, Spanish > and Portuguese versions of this disclaimer. > > > This message and any attachment are confidential and may be privileged or > otherwise protected from disclosure. If you are not the intended recipient, > you must not copy this message or attachment or disclose the contents to any > other person. If you have received this transmission in error, please notify > the sender immediately and delete the message and any attachment from your > system. Merck KGaA, Darmstadt, Germany and any of its subsidiaries do not > accept liability for any omissions or errors in this message which may arise > as a result of E-Mail-transmission or for damages resulting from any > unauthorized changes of the content of this message and any attachment > thereto. Merck KGaA, Darmstadt, Germany and any of its subsidiaries do not > guarantee that this message is free of viruses and does not accept liability > for any damages caused by any virus transmitted therewith. > > Click http://www.merckgroup.com/disclaimer > http://www.merckgroup.com/disclaimer to access the German, French, Spanish > and Portuguese versions of this disclaimer.
Dec 05, 2019 Emeric Sibieude Block SIZE with SCM
Dec 05, 2019 Jakob Ribbing Re: Block SIZE with SCM
Dec 09, 2019 Emeric Sibieude RE: Block SIZE with SCM
Dec 09, 2019 Jakob Ribbing Re: Block SIZE with SCM
Dec 09, 2019 Rikard Nordgren Re: Block SIZE with SCM