UUPharmacometrics / PsN

Perl-Speaks-NONMEM
https://uupharmacometrics.github.io/PsN
GNU General Public License v2.0
65 stars 19 forks source link

Problem with PsN_sir_plots.R (sir-5.2.6) #214

Open sundav05 opened 3 years ago

sundav05 commented 3 years ago

Hi,

One of our SME ran sir-5.2.6 with the arguments shown below.

sir-5.2.6 run2.mod -samples=1000,1000,2000,2000 -resamples=100,250,1000,1000 -rplots=2 -clean=1 -dir=sir2

sir runs fine but the plotting fails with the following error.

raw.results.file <- c() for (i in ALL.RAWRESFILES) {

  • rawres.cur <- read.csv(paste(i,sep=""))
  • rawres.cur$ITERATION <- which(ALL.RAWRESFILES==i)
  • raw.results.file <- rbind(raw.results.file,rawres.cur)
  • } Error in read.table(file = file, header = header, sep = sep, quote = quote, : more columns than column names Calls: read.csv -> read.table Execution halted

This is from PsN internal "R" code so it is not stemming from a user error. Examination of the iteration output files reveals that header line is 35 column wide, the "0" iteration line is 84 columns wide and the rest of the rows are all 80 columns wide. I have included below the 3 lines from raw_results for your review. Can you please confirm if this is a known bug and whether there are any remedial steps we could apply. Thanks!

$ head -3 raw_results_run2.csv "sample.id","model","problem","subproblem","covariance_step_run","minimization_successful","covariance_step_successful","covariance_step_warnings","estimate_near_boundary","rounding_errors","zero_gradients","final_zero_gradients","hessian_reset","s_matrix_singular","significant_digits","condition_number","est_methods","model_run_time","subprob_est_time","subprob_cov_time","ofv","deltaofv","likelihood_ratio","relPDF","importance_ratio","probability_resample","resamples","sample_order","1 [CL] L/d","2 [V1] L","3 [V2] L","4 [Q] L/d","5 KA","6 F","Proportional err","1 CL","OMEGA(2,1)","2 V1","3 V2","4 Q","5 KA","6 F3","SIGMA(1,1)","se1 [CL] L/d","se2 [V1] L","se3 [V2] L","se4 [Q] L/d","se5 KA","se6 F","seProportional err","se1 CL","seOMEGA(2,1)","se2 V1","se3 V2","se4 Q","se5 KA","se6 F3","seSIGMA(1,1)","shrinkage_eta1(%)","shrinkage_eta2(%)","shrinkage_eta3(%)","shrinkage_eta4(%)","shrinkage_eta5(%)","shrinkage_eta6(%)","shrinkage_iwres(%)","EI1","EI2","EI3","EI4","EI5","EI6","EI7","EI8","EI9","EI10","EI11","EI12","EI13","EI14","EI15" 0,input,1,1,1,1,1,0,0,0,NA,NA,NA,0,NA,NA,SAEM_I-IMP_I,0:03:27,160.69,34.54,-1111.613,0,NA,NA,NA,NA,NA,NA,0.28,3.76,6.1,0.153,0.166,0.531,0.259,0.593,0.314,0.17,0.0225,1.73,0.366,0.00578,1,0.0431,0.257,0.508,0.0911,0.0238,0.00497,0.0205,0.282,0.0858,0.0384,0,1.76,0.144,0.00863,0,NA,NA,NA,NA,NA,NA,NA,4.33E-04,4.32E-02,5.58E-02,8.62E-02,1.39E-01,2.15E-01,4.42E-01,5.84E-01,1.03E+00,1.14E+00,1.58E+00,1.69E+00,5.99E+00,Elapsed,postprocess,time,in,seconds:,2.80 1,1,1,1,0,1,0,0,0,0,NA,NA,NA,0,NA,NA,IMP_I,2:28:35,91.23,0,-1104.7686807762577,6.84431922374233,0.0326418650054516,0.025535152204715,1.27831096301139,3.52215640710915e-05,1,721,0.272919,3.72568,5.81539,0.158576,0.156868,0.532612,0.252479,0.535979,0.278173,0.156116,0.0225,3.64999,0.37575,0.00713517,1,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,NA

rikardn commented 3 years ago

Thanks for reporting this. It seems to be a bug in the generation of the raw results file. The message "Elapsed postprocess time..." shouldn't be in the table. A workaround would be to replace the read.csv in the R script template with some code that reads the csv without complaining about the imbalance in the number of columns.