Dear all, Meanwhile we further investigated the issue. We might need some information and advice from Robert Bauer
Two issues have been identified. ->MH sampling in general -> leads to some instability in the OBF if more nodes are used ->More important: the use of a sigma as proposed in NONMEM 7.3.0:-. .......---- SIGMA(x,x) has been allowed to be used on the right hand side of equations in the control stream file. In this model, we followed the recommendation for BLQ values by designating the GRD appropriately:. $EST METHOD=IMP LAPLACE INTERACTION CTYPE=3 NOHABORT GRD=SN(1) PRINT=1 In the test cases this leads to different OBF in case of parallelization's (in the first iterations in some models OBF drastically increase) independent of the use of parse_type=1 und ranmethod=p.. For further investigations, we would need the exact implementation in NONMEM. For the moment we will stay away from using Sigma on the right hand side with IMP and BLQ values. Any information and advice would be highly appreciated Best wishes Dirk -----Ursprüngliche Nachricht----- Von: owner-nmus...@globomaxnm.com [mailto:owner-nmus...@globomaxnm.com] Im Auftrag von Dirk Garmann Gesendet: Dienstag, 20. September 2016 20:05 An: Leonid Gibiansky; nmusers@globomaxnm.com Betreff: AW: AW: [NMusers] IMP and parallelisation Dear Leonid Still in the evaluation phase. I thought it was already finished but still running (long run times). I will keep you informed if the results are identical Best Dirk -----Ursprüngliche Nachricht----- Von: Leonid Gibiansky [mailto:lgibian...@quantpharm.com] Gesendet: Dienstag, 20. September 2016 15:54 An: Dirk Garmann; nmusers@globomaxnm.com Betreff: Re: AW: [NMusers] IMP and parallelisation Hi Dirk, What do you mean "does not solve the issue"? Were the results identical with different number of nodes or not? Thanks Leonid On 9/20/2016 9:47 AM, Dirk Garmann wrote: > Thank you Leonid, > We have tried RANMETHOD=P, which is an interesting possibility. > Unfortunately this does not solve the issue. We will further evaluate if the > information from all nodes is used for the population update. > Any further hints are highly welcome > > > Best > Dirk > > -----Ursprüngliche Nachricht----- > Von: Leonid Gibiansky [mailto:lgibian...@quantpharm.com] > Gesendet: Montag, 19. September 2016 22:26 > An: Dirk Garmann; nmusers@globomaxnm.com > Betreff: Re: [NMusers] IMP and parallelisation > > It is a good idea to use RANMETHOD=P at estimation step; then the > results should be identical independently of the number of nodes and > computer load. > > Concerning specific behavior .. looks strange. I would try to start from > the initial values of the model with the lowest OF and see what happens. > > Thanks > Leonid > > > On 9/19/2016 1:29 PM, Dirk Garmann wrote: >> Dear nmusers. >> >> During a popPK analysis using the M3 method and IMP we observed an >> unexpected behavior and would be interested if anyone else observed the >> same and can provide guidance/explanations. >> >> >> >> The IMP produces "strange" results in cases requiring a parallelization. >> >> We observed a general (and strong) trend that with increasing number >> of nodes the OBF increases (!) which in my opinion is unexpected if the >> number of samples in MC is sufficiently large. >> >> >> >> The initial settings have been as follows: >> >> Parse Type 1 >> >> >> >> $EST METHOD=IMP INTERACTION LAPLACIAN EONLY=0 ISAMPLE=300 NITER=1000 >> CTYPE=3 NOABORT GRD=SN(1,2) NOTHETABOUNDTEST PRINT=1 >> >> $EST METHOD=IMP INTERACTION NOABORT GRD=SN(1,2) EONLY=1 ISAMPLE=3000 >> NITER=30 PRINT=1 >> >> >> >> With 1 node the OBF decreased to ~- 1400 >> >> Using 16 nodes the OBF stabilized at ~ 1000 >> >> In both cases the OBF does not fluctuate much after 100 interations >> (monitoring of EM step) and seems to be stable (no clear hint for a >> local minima). >> >> Interestingly the estimated residual error is higher using 1 node. With >> 16 nodes the variability seems to be shifted to the ETAS. >> >> >> >> This behavior might be a concern for a covariate analysis using IMP >> >> Our first assumption was that we need to increase iSAMPLE in the EM >> step, since a different seed might be used for each node. However even >> increasing ISAMPLE to 3000 in the first step did not change the results >> much. >> >> My guess is that it points in the direction of how population values are >> updated, but I am not an expert in the implementation of IMP in NONMEM >> >> >> >> We would be highly interested in any guidance and explanation. >> >> >> >> Many thanks in advance >> >> >> >> Dirk >> >> >> >> Freundliche Grüße / Best regards, >> >> >> >> Dirk Garmann >> >> Head Quantitative Pharmacology >> >> >> >> >> >> Bayer Pharma Aktiengesellschaft >> >> BPH-DD-CS-CP-QP, Quantitative Pharmacology >> >> Building 0431, 322 >> >> 51368 Leverkusen, Germany >> >> >> >> Tel: +49 202 365577 >> >> Fax: >> >> Mobile: +49 175 3109407 >> >> E-mail: _dirk.garmann@bayer.com_ >> >> Web: _http://www.bayer.com_ >> >> >> >> Vorstand: Dieter Weinand, Vorsitzender | Christoph Bertram >> >> Vorsitzender des Aufsichtsrats: Hartmut Klusik >> >> Sitz der Gesellschaft: Berlin | Amtsgericht Charlottenburg, HRB 283 B >> >> >> >> >> >> >> >
