Peptide QSAR — Advanced Options - PLS Dialog Box

Advanced Options - PLS Dialog Box Features

• Maximum number of PLS factors box
• Autoscale X variables option
• Stop adding PLS factors when standard deviation of the regression drops tooption and text box
• Eliminate X variables with t-value < text box

Maximum number of PLS factors box

Specify the maximum number of PLS factors to use in the regression model. Regression models are built for increasing numbers of PLS factors up to this number. The maximum number that can be used is limited by the number of descriptors, which is 3 times the number of residues for the zvalue set, 5 times the number of residues for the ezvalue set, and 10 times the number of residues for the dpps set. It is rarely useful to build models with more than a few PLS factors, as models with a large number tend to be overfit. You should examine the statistics, particularly the stability and Q², to determine how many PLS factors to use in the model you choose for application to new systems.

Autoscale X variables option

Scale the X variables by dividing the values of each property by the standard deviation in the value of the property.

Stop adding PLS factors when standard deviation of the regression drops to option and text box

Select this option to stop adding PLS factors when the standard deviation of the regression drops below the value specified in the text box. The value in the text box should be related to the error in the measured values that you are fitting. Using this option could result in fewer PLS factors than the number specified in the Maximum number of PLS factors box, but adding more factors may not yield any improvement in the model.

Eliminate X variables with t-value < text box

Eliminate X variables whose t-value is less than the value given in the text box. The t-value is the ratio of the coefficient of the variable in the fitted model to the standard error of the model. Small t values indicate that the variable is not contributing significantly to the model.