Bio-Rad Bio-Plex Software® Upgrades and Conversions - Manual (Page 185)

Standard Curve
173
Linear vs. Logistic Regression Methods
In terms of performance, the primary difference between linear and logistic
regression models is that the linear range in which concentrations can be
accurately predicted is much smaller. In most cases, the overall response of
the assay is best modeled using a logistic fit and the Logistic-5PL model will
yield the best results. For more information, see Principles of Curve Fitting
from the website (
www.bio-rad.com ) or from Technical Support.
Nevertheless, in certain cases, you may want to use linear regression to
analyze the data. For most cytokines, the linear portion of the curve is well
within the biological range of concentrations expected for patient samples.
Thus, the linear regression may be useful for serum. The main advantage of a
linear regression is that fewer points (as few as 2) can be used.
Copying the Standard Curve
To copy a bitmap image of the curve to the Windows clipboard, right-click in
the Standard Curve window, and select Copy Graph to Clipboard. You can
then paste this image into your spreadsheets or other documents, using the
Paste command in your other applications.
Figure 156. Copying graph to clipboard
Exporting the Standard Curve
To export a standard curve to Excel, follow the steps below.
1. Select the Export Standard Curve icon from the toolbar.
2. Select your choice of exporting an individual analyte, or all analytes, in
the Export Standard Curve dialog. Legends, the standard curve
equation, fit probability and residual variance display below the chart.
BioPlex_6.book Page 173 Friday, September 23, 2011 11:55 AM