Phoenix WinNonLin
Jump to navigation
Jump to search
Connecting
Instructions
Tentative Pheonix WinNonLin Project
Task #1-Learn WinNonLin
Click for instructions here. You can stop at "Execute noncompartmental analysis"
- Make an MS Word docx document called Task01-Learning_Phoenix_WinNonLin.docx
- On the first page, describe what each of the abbreviations mean.
- Open Phoenix WinNonLin
- Create a new project called Task01-Learning_Phoenix_WinNonLin.phxproj
- Import dataset based on your student number such as Task01-Student01.dat
- Make a XY plot of your data (e.g., Task01-Student01.dat)
- Add the XY plot to one of the pages of your MS Word docx: Task01-Learning_Phoenix_WinNonLin.docx
- Make a table of your data.
- Add the table to one of the pages of your MS Word docx: Task01-Learning_Phoenix_WinNonLin.docx
- In Task01-Learning_Pheonix_WinNonLin.docx, and answer the following questions on the last page (Simple definitions please. Does not have to be complicated mathematics.):
- What is the difference between mean, harmonic mean, and geometric mean?
- What is the difference between standard deviation, standard error and pseudo standard deviation?
- What does coefficient of variation (CV%) mean in the context of this data?
- What does interquartile range (IQR) mean in terms of pharmacokinetics?
- Send Task01-Learning_Pheonix_WinNonLin.docx and Task01-Learning_Pheonix_WinNonLin.phxproj to me. You can email me directly from the remote desktop.
Tips, Additional Instructions, and Information
- The instructions are pretty good. The task takes about 1-2 hours.
- I recommend doing the example first.
- Change "Map Subject to the Stratification Row context." to "Map Subject to the Stratification Column context." because it is easier to read in MS Word 2013.
- The path is wrong in the instructions. The path should be C:\Program Files (x86)\Pharsight\Pheonix\application\Examples\Bguide1.dat
- This might be confusing, use the Microsoft Word 2013 that is on the Remote Desktop
- Right click on Figures to Copy them into Microsoft Word 2013
- Right click under Result Tables → Tables on the right. Then click Export... and save it somewhere where you can find it in html format. The default is table.html.
- Open the html file as a separate document and copy the table into the main document.
- You can find a lot of questions by going into Google and typing, "meaning of standard error" for the answer to question #2.
- If you are using Windows 10, you can paste directly back and forth between the Remote Desktop and your computer.
- Useful Links
Student Data
- Task01-Student01.dat
- Task01-Student02.dat
- Task01-Student03.dat
- Task01-Student04.dat
- Task01-Student05.dat
- Task01-Student06.dat
- Task01-Student07.dat
- Task01-Student08.dat
- Task01-Student09.dat
- Task01-Student10.dat
- Task01-Student11.dat
Task #2-Dr. Roberts' Attempt to Fit Pharmacokinetic Parameters
- I like very detailed instructions because the instructions make it easier for me to go back where I was.
- Make an MS Word docx document called Task02-Fit_Pharmacokinetics.docx
- Create a new project called Task02-Fit_Pharmacokinetics.phxproj
- Click
- File → New Project or Control + N
- I named the file "Dr Roberts PK Project"
- Downloaded Task02-Student10.csv, since I am Student10. You will load the appropriate number for you.
- Right Click on Data, which is on the left side and right click import and open the data that you downloaded.
- Select the Radio Button for Has units row. I missed it the first time.
- Click
- Right-click Workflow in the Object Browser and select New > Plotting > XY Plot.
- Double click on the XY Plot
- Drag the Task02_Student10 worksheet to the XY Plot window. This can be a little tricky. It might take you a couple tries.
- Leave Subject mapped to the None context. Map Time to the X context. Map Conc to the Y context.
- Click the
- The amplitude of my specific example is 0.023 ng/ml, which is 0.023 μg/L.
- Dosage guess if there is no volume distribution (V): Now, let us calculate the systemic circulation dosage assuming that there is no distribution in the tissues. Therefore the apparent volume distribution (V) = the volume in the systemic circulation (V_{v}) = 5 L.
- Av = dosage is the systemic circulation
- Therefore C = A_{v}/V_{v} > C x V_{v} = A_{v}
- 0.023 μg/L * 5 L = 0.115 μg
- Dosage guess if there is volume distribution (V): Let us pick an intermediate volume distribution like 200 L.
- Av = dosage is the systemic circulation
- Therefore C = A/V > C x V = A
- 0.023 μg/L * 200 L = 4.6 μg
- In the Options tab below the plot, select Axes > Y from the menu tree.
- Select the Logarithmic option button in the Scale area. Leave the logarithmic base set to 10. This makes it easier to make estimates of the absorption rate constant (k_{a}) and the elimination rate constant (k_{e}).
- Right Click > New > Modeling > Least-squares Regression Models > PK Model
- I dragged Task02_Student10 data into the PK model
- Map Subject to the Sort context. Leave Time mapped to the Time context. Map Conc to the Concentration context.
- In the Model Selection tab below the Setup panel, check the Number 3 model checkbox.
- Select the Dosing panel in the Setup tab. Check the Use Internal Worksheet checkbox.
- Click
- No distribution of drug: Put in a time of 0 and a dose of 0.115.
- In the Weighting/Dosing Options tab below the Setup panel, type ug in the Unit field.
- In the Parameters option menu, select the User Supplied Initial Parameter Values option.
- Select the Initial Estimates tab, check the Internal Worksheet Box and click OK when the dialog comes up.
- Go to the XY Plot and select a point that is half way on the absorption phase. It should look something like:
- The two values are 0.4 and 0.013, which represent the time and the concentration. An approximate absorption rate constant (k_{a}) = 1/0.4 = 2.5 hr^{-1}.
- Go to the XY Plot and select a point that is half way on the elimination phase. It should look something like:
- I could not get the numbers to show up, but the half way point is about 12.5 hrs. An approximate elimination rate constant (k_{e}) = 1/12.5 = 0.08 hr^{-1}.
- Now, let us go back to the model:
- K01 is the k_{a} and K10 is the k_{e}.
- Go back and Select the Initial Estimates tab.
- For K01, I put 2.5 hr^{-1} and for K10, I put 0.08 hr^{-1}, but what is the V_F term? The V_F term is in units of μg/(ng/mL). That is very strange. Let us simplify.
- μg/(ng/mL) = μg/(μg/L)> μg cancel out, which leaves us liters. Why don't they simplify in the first place? I have no idea. In the original example, they have 0.25 L, but let us try 5L like the volume of the systemic circulation. That would be the box with 1 in it above.
- For K01, I put 2.5 hr^{-1}, for K10, I put 0.08 hr^{-1} and for V_F, I put 5 L for the blood volume in the systemic circulation. This is what I have:
- Press the execute button in WinNonLin .
- Make a section in Task02-Fit_Pharmacokinetics.docx called, No Drug Distribution
- Make a section in Task02-Fit_Pharmacokinetics.docx called, 200 L Volume Distribution
- Change the dosage to 4.6 μg and the V_F to 200 L. Rerun the simulation. You will have to go to the setup section to change the parameters.
- Put the Final Parameters on one page. You can copy a screenshot or you can make a table.
- Put the Observed versus Predicted on one page.
- Make a section in Task02-Fit_Pharmacokinetics.docx called, More Realistic Example.
- Your patient is taking a 1 mg of a drug orally. The C_{max} (see above) is approximately 0.023 μg/L.
- Therefore, the approximate apparent volume distribution will be related to C=A/V > V = A/C. 1,000 μg/0.023 μg/L = 43,478 L (It can be that large. Might be stuck in adipose tissue)
- Change the dosage to 1000 μg and the V_F to 43478. Rerun the simulation. You will have to go to the setup section to change the parameters.
- Put the Final Parameters on one page. You can copy a screenshot or you can make a table.
- Put the Observed versus Predicted on one page.
- Your patient is taking a 1 mg of a drug orally. The C_{max} (see above) is approximately 0.023 μg/L.
- Notice that different parameters can fit the same curve.
- Send Task02-Fit_Pharmacokinetics.docx and Task02-Pharmacokinetics.phxproj to me
Tips and Notes
- The path of the example is wrong in the instructions. The path should be C:\Program Files (x86)\Pharsight\Pheonix\application\Examples\
Student Data
- Task02-Student01.csv
- Task02-Student02.csv
- Task02-Student03.csv
- Task02-Student04.csv
- Task02-Student05.csv
- Task02-Student06.csv
- Task02-Student07.csv
- Task02-Student08.csv
- Task02-Student09.csv
- Task02-Student10.csv
- Task02-Student11.csv
Task #3-Non-Compartment Analysis
Click for instructions herePDF Version
- Make an MS Word docx document called Task03-Non-Compartment_Analysis.docx
- On the first page of Task03-Non-Compartment_Analysis.docx, put the following parameters:
- Tau, Rsq_adjusted, MRTINF_pred, TimeLow, TimeBetween, TimeHigh, TimeInfBetween, Lamda Z
- Explain them in simple language. You can usually find them through a Google search.
- Open Phoenix WinNonLin
- Create a new project called Task03-Non-Compartment_Analysis.phxproj
- Import dataset based on your student number such as Task03-Student01.dat.
- Put a table of the results of one of the patients onto a page in Task03-Non-Compartment_Analysis.docx
- Put the NCA plot for the same patient onto a page of Task03-Non-Compartment_Analysis.docx
- Put the Final Parameters Pivoted onto a page of Task03-Non-Compartment_Analysis.docx
- Put the Ratios Differences onto a page of Task03-Non-Compartment_Analysis.docx
- In Task03-Non-Compartment_Analysis.docx, describe what your results mean in a descriptive language.
- Send Task03-Non-Compartment_Analysis.docx and Task03-Non-Compartment_Analysis.phxproj to me.
Tips and Notes
- Example data does not work. Your student data should work the same as the example.
- Read the instructions very carefully. This section took me a while to get through.
Student Data
- Task03-Student01.dat
- Task03-Student02.dat
- Task03-Student03.dat
- Task03-Student04.dat
- Task03-Student05.dat
- Task03-Student06.dat
- Task03-Student07.dat
- Task03-Student08.dat
- Task03-Student09.dat
- Task03-Student10.dat
- Task03-Student11.dat
Task #4-Bioequivalence
Click for instructions herePDF Version
- Make an MS Word docx document called Task04-Bioequivalence.docx
- On the first page describe the following abbreviations in easy to understand language:
- ln(AUClast), Hypothesis, Numer_DF, Denom_DF, F_Stat, P_value, CI_90_Lower, CI_90_Upper
- In simple language, describe what a 2x2 and a 2x4 crossover study.
- Open Phoenix WinNonLin
- Create a new project called Task04-2x2-Bioequivalence.phxproj
- Import dataset based on your student number such as Task04-2x2-Student01.csv.
- Put a copy of the Average Bioequivalence worksheet on a page of Task04-Bioequivalence.docx
- Label and underline with 2x2 Crossover
- Put the Partial Tests and Sequential Tests worksheets on a page of Task04-Bioequivalence.docx
- Label and underline with 2x2 Crossover
- Create a new project called Task04-2x4-Bioequivalence.phxproj
- Import dataset based on your student number such as Task04-2x4-Student01.csv
- Put a copy of the Average Bioequivalence worksheet on a page of Task04-Bioequivalence.docx
- Label and underline with 2x4 Crossover
- Put the Partial Tests and Sequential Tests worksheets on a page of Task04-Bioequivalence.docx
- Label and underline with 2x4 Crossover
- Put the Population Individual worksheet on a page of Task04-Bioequivalence.docx
- Label and underline with 2x4 Crossover
- In Task04-Bioequivalence.docx, describe what your 2x2 and 2x4 crossover study mean in a descriptive language.
Tips and Notes
- The path of the example is wrong in the instructions. The path should be C:\Program Files (x86)\Pharsight\Pheonix\application\Examples\
- Go through the example first.
- For the 2x2 Crossover, the Random Effects model is supposed to be Subject(Sequence) and the Type is Variance Components. When it was not set to this, WinNonLin crashed.
Student Data
2x2 crossover
- Task04-2x2-Student01.csv
- Task04-2x2-Student02.csv
- Task04-2x2-Student03.csv
- Task04-2x2-Student04.csv
- Task04-2x2-Student05.csv
- Task04-2x2-Student06.csv
- Task04-2x2-Student07.csv
- Task04-2x2-Student08.csv
- Task04-2x2-Student09.csv
- Task04-2x2-Student10.csv
- Task04-2x2-Student11.csv
2x4 crossover
- Task04-2x4-Student01.csv
- Task04-2x4-Student02.csv
- Task04-2x4-Student03.csv
- Task04-2x4-Student04.csv
- Task04-2x4-Student05.csv
- Task04-2x4-Student06.csv
- Task04-2x4-Student07.csv
- Task04-2x4-Student08.csv
- Task04-2x4-Student09.csv
- Task04-2x4-Student10.csv
- Task04-2x4-Student11.csv
Questions from Students
How do you guess the following parameters tvKa, tvV, tvKe?
- A oral dosage will have and absorption phase and an elimination phase. The tvKa and the tvKe are pretty easy to estimate. The time where the amplitude of the absorption phase is half. Let us say that it is 0.2 hrs. Then a good guess for the tvKa would be 1/0.2 hrs or 5 hr^{-1}. Let us say that the half amplitude of the elimination phase is 2 hours. Then a good guess for the tvKe would be 1/2 hrs or 0.5 hr^{-1}.
- With the third parameter, I would use the slider until it close.
Pheonix WinNonLin | Classic WinNonLin | Likely Meaning | Guess (Y/N) |
---|---|---|---|
tvKa | K01 | absorption rate constant (K_{A}) | Y |
tvV | V1_F | Volume of the central compartment? | Y |
tvKe | K10 | elimination rate constant (Ke) | Y |
tvK12 | K12 | rate constant away from the central compartment to a peripheral compartment | Y |
tvK21 | K21 | rate constant from the peripheral compartment to the central compartment | Y |