2020-05-09 tc1510_M1SD CTOLDesti parameter not generated #233
Comments
|
Opened in error. CTOLDesti was included in tc1510_M1SD MCT incorrectly. Will be removed from MCT and this issue can be closed. |
|
This is reopened because it was intended to include CTOLDesti for Single dose models as per the CE specifications documentation for CTOLDesti: Single Dose Model (M1,M3) For Single Dose Model, the concentration for CTOLDESTi will be imputed to a value of zero, 0. Capturing the value of CTOLDESTi permits recording what was used for AUC computations where the data point at time of dosing is missing. |
opennca
added a commit
that referenced
this issue
Aug 27, 2020
The following scope items were implemented: 3.0a Derived/Secondary Parameter Computations 3.0b Confirm/Test INCLUDEINTERPOLATION functionality for Partial Area Computations 3.0c Confirm/Test INCLUDEEXTRAPOLATION functionality for Partial Area Computations 3.0e Version Identification 3.1a "Performance improvement Computation Engine Performance Targets 1. Computation Engine execution virtual compute call (one profile; 20 nominal time datapoints) <3 sec 2. Computation Engine execution virtual compute call (one profile; 20 nominal time datapoints) <15 sec with optimization 3. Computation Engine execution compute all profiles (100 profiles; each 20 nominal time datapoints) <1 min 4. Computation Engine execution compute all profiles (100 profiles; each 20 nominal time datapoints) <2 min with optimization Elapsed time is defined as the time following initiation of the run_computation function call to completion and closing of all output from that call." 3.1b 2. KEL Optimization logic could be optimized to see how much time could be saved. 3.2 "v1.defect#1: Units and scaling for weightnormalized parameters Github 247: 2020-05-14 tc1513 M2SD weight-normalised parameters provided when NORMBSU is not valid Github 232: 2020-05-08 tc1513 M2SD unit_conversion warning messages falsely lists columns as missing from mct/data" 3.3 "Unit conversion of TIME to minutes is not applied. Parameter unit label TIMEU is MIN in the output but parameter values are still in hours. Add TIME UNIT of ""H"" to option of ""HR"" in unit_conversion.R to avoid warning messages that ""H"" is inappropriate" 3.4 FLGACCEPTAUCTAU is not applied as intended when KEL is not estimated. 3.5 AUMCTAU is inappropriately reported when AUCTAU is not 3.6 Plasma/serum TLAG is not correct 3.7 "LOG AUC parameters have not been validated LOG AUC method crashes" 3.8 "Data handling for AUCTAU using actual times with more than one dosing interval may not be exactly as intended. Github 244: 2020-05-13 tc158 M1SS AUCTAU2 not matching partial AUC for same time period" 3.9a "v1.defect#3: Potential invalid parameter results when using unrecognized units (eg, DPM). Github 225: 2020-04-07 tc157 incorrect DOSE units doesn't always trigger warning and continues to produce parameter data" 3.9b Need to generate a warning when Dose / Time / conc are inconsistent. 3.1 2020-05-09 tc1510_M1SD CTOLDesti parameter not generated #233 3.11 Add capability to handle dose units by weight (e.g. mg/kg) 3.12 "Scope #2: Update CE spec and methods for CMINDN (SS models) and for CENDINF/TENDINF (M3 model) Scope #21: tc113 M3SD doesn’t generate CMINDN" 3.12b Scope #2: Update CE spec and methods for CENDINF/TENDINF (M3 model) 3.13 AUCINFP and other ‘P” parameters computed with no KEL 3.14 tc125 M1SS fails to find TAU/TOLD values in either input concentration dataset or flags dataset 3.15 2020-04-17 tc1518_M4SS - concerning warning message from run_computation 3.16 "Inconsistent data handling with negative times (this was present in v1 but was missed in the testing reviews)" 3.18 "Execution fails with missing data (“NA” data point with no concentration result, as opposed to below limit of quantification or no record) at end of profile. Github 238: 2020-05-12 tc1510_M1SD-1 testcase variant fails with “interpolate_lin” error" 3.19 AUMC and MRT in steady-state models with more than one dosing interval. 3.21 "Github 201: tc155 M1SS computation engine doesn't generate error when duplicate time values appear with different concentration values v1.defect#12: Ambiguous warning message for duplicate times" 3.22 "FLGEXKEL computation error when using OPTIMIZEKEL in five testcases introduced with a6cb6ff OPTIMIZEKEL results not consistent with generated FLGEXKEL values, generated parameters nor FLGACCEPTKELCRIT settings 2020-05-14 tc119 M1SD KEL NA for unknown reason when OPTIMIZEKEL == 1 for two profiles FLGNOCMAX isn't properly respected in all cases." 3.23 "2020-04-02 tc1572 M4SS - CONC and CONCTIME, parameter outputs not provided in results from run_computation 2020-05-12 tc1518_M4SS Model 4 parameter results do not include CONC/CONCTIME outputs Decision is to REMOVE CONC/CONCTIME outputs from M1SD and continue to exclude CONC/CONCTIME from M4SS. The work estimate for this must be sure to account for this. To be clear, this was changed to reflect NOT adding the missing CONC/CONCTIME parameter outputs to M4SS results datasets but remove from M1SD. " 3.24 "MRTEVIFP1 is not calculated correctly. (Not identified previously due to v1.defect#8)" 3.25 "Emesis flag – issue noted in CE spec review Note that the work effort is to adjust the CE spec (Pfizer effort) and Rudraya is to ensure that the handling of the data associated with Emesis flag is then consistent with the updated specification." 3.26 "No urine parameters generated with actual times if start time of first interval is negative. Observed in M4SD. Actual times not tested with M4SS." 3.27 Urine TLAG not computed correctly (M4 models). The midpoint time of first interval is used instead of 0. 3.29 2020-04-03 tc111 M4 RATEN returning NaN instead of NA when mct$ACTENDTIME <= mct$ACTTIME 3.3 2020-04-03 tc111 M4 TMAXRATE selecting wrong interval when RATE is NaN 3.32 "When NORMBS and NORMBSU entries in MCT are missing or not correctly specified (set but are missing in dataset for example), Body Size normalized parameters such as CLFOW, CLFPW,CLFTAUWi,CLOW,CLPW,CLTAUWi,VSSOW,VSSOWi,VSSPW,VSSPWi,VZFTAUWi,VZOW,VZPW,VZTAUWi,VZFOW, VZFPW should not be attempted to be computed. Github #252" 3.33 AUCTAU for single dosing interval – add Github issue and testcase from the original
|
This issue was addressed in commit 472a0a2. While CTOLDesti is still reported as NA for some cases, this is because TOLD is not specified for SD models. After implementing CTOLDest should be estimated at 0 if concentration at TOLD is missing, CTOLDesti is still reporting NA, but correctly (since 0 is part of the profile for all subjects). ## Run tc1510 M1SD
paste(testcase.id, mct$MODEL, mct$DOSINGTYPE, mct$AUCMETHOD, mct$TIME, sep = ".")
#> "tc1510.M1.SD.LINLOG.Nominal"
results_list <- run_computation(data = d, map = mct, flag = flags,
parameterset = parameterset)
#> ... computation engine messages ...
r <- results_list$data_out
r$CTOLDest1
#> [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 NA NA NA
#> [41] NA NA NA NA NA NA NA NA
## Run tc1510 M1SD but with all TIME == 0 removed
results_list <- run_computation(data = d, map = mct, flag = flags,
parameterset = parameterset)
#> ... computation engine messages ...
r <- results_list$data_out
r$CTOLDest1
#> [1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Tested with commit b76b846
The text was updated successfully, but these errors were encountered: