StatLIA® measures relative potency and parallelism even for bioassays that yield difficult, ill-behaved data.
StatLIA® Parallelism measures the biological similarity between two molecules quantitatively, where parallelism and non-parallelism can be directly compared. StatLIA® Parallelism using  methodology provides results that are:
Click to enlarge |
- Consistent
Determine potency even for assays with high variation using StatLIA’s correct weighting
|
Click to enlarge |
- Accurate
Improve the accuracy of full curve parallelism analysis with StatLIA’s TrueFit™ 5PL
|
Click to enlarge |
- Reliable
Measure parallelism and nonparallelism directly, accurately, every time using method
|
Presentations and Posters
U.S. Pharmacopeia Bioassay Workshop, August 2008
Download Brendan CTO Presentation: "Nonlinear Systems: A Comparison of Approaches to Similarity Testing."
Link to all Presentations at USP Bioassay Workshop.
IBC Biological Assay Development & Validation Conference, May 2008
Download MERCK Presentation:
(MERCK Uses STATLIA Parallelism Metric to Set Acceptance Criteria)
Download Presentation: "Comparison of Methods Used to Measure Parallelism"
Download Poster: "Comparison of Methods Used to Measure Parallelism"
Data: (Request Data Used In Parallelism Comparison Study)
Reference
Measuring Parallelism, Linearity and Relative Potency in Bioassay and Immunoassay Data, Gottschalk PG, Dunn JR, Journal of Biopharmaceutical Statistics, 2005; 3:437-463. (Free Reprint Available)
| |
|
Key Parallelism Metrics
- Probability: Probability the two curves are parallel based on the Chi-Square statistic
- Statistic: Direct measure of the amount of nonparallelism between two curves
- Relative Potency: Relative potency between the two curves
- Low Confidence Limit: Lower 95% confidence limit for the relative potency
- High Confidence Limit: Upper 95% confidence limit for the relative potency
- Confidence Limit Ratio: Ratio of the logs of the upper and lower confidence limits
- IC25, IC50, IC75: Percent concentration of the maximum response calculated from the unknown dilution curve
- Dil25, Dil50, Dil75: Equivalent dilution factor of the inhibitory concentration, e.g. (Dil*Conc)/IC50
- Fit Probability: Measurement of how well the computed curve fit the observed data points
- Limits of Quantitation and Detection: Limits of Quantitation and Detection based on the observed data points
- Dilution CV's: Coefficient of variation of all reportable concentrations in a dilution series
- Download Parallelism Poster PDF File
- Download Parallelism PDF File
|
| |
Full Logistic Curve Parallelism Method
- Use a powerful weighted 5 parameter or 4 parameter logistic model to fit the curves
- Measure the reliability of the standard curve
- Measure the reliability of the unknown dilution curve
- Measure the parallelism between the two curves using the Statistic and Probability
- Download Parallelism Fact Sheet PDF file
|
|
