QUICK LINKS
Free Trial
Request Pricing & Information
Get Technical Support
Email or Call 1-888-589-4629
Never Let the Important Become Urgent: A reflection on the genetics supply chain and our need to increase value to the end patient
» Read blog post
Regression Module
Perform advanced linear and logistic regression, stepwise regression, and permutation tests with numeric variables and recoded genotypes.
Check out the following tutorial on logistic regression, which covers controlling for confounding variables, model comparison, and stepwise regression.
Or find out how to set up a linear regression model on Our 2 SNPs...®.
Full vs. Reduced Models
Sometimes it is desired to correct for binary, continuous, or categorical variables, otherwise known as “covariates”. These covariates, or first-order interactions between covariates, may be influencing the dependent variable response. Correcting for covariates allows you to see specifically what effects there are on the remaining variables. To do this, first a linear regression equation, which includes only the dependent and the reduced model covariates, is calculated (the “reduced model”). Next, a linear regression which includes all variables and full model covariates is calculated (the “full model”). The significance of the full versus the reduced model is calculated with an F-test.
Linear and Logistic Regression
A number of regression options are available for case/control and quantitative traits, using phenotypic, CNV, and genotypic covariates. Choose among linear or logistic regression regressing once on each numeric column in your spreadsheet, using a moving window of regressors, or on selected covariates only. By using the Recode Genotypes option you can also perform regression on genotype columns. Also available is the ability to include first-order interactions.
Stepwise Regression
Stepwise regression is an automated procedure that sequentially adds or removes predictor variables to or from a regression model, helping to identify subsets of significant predictors. Backward elimination starts with all of the full-model covariates and removes the least significant covariate until removing any covariates would be more significant than the stepwise p-value cut-off specified. Forward selection selects the most significant covariate and keeps adding the next most significant covariate until adding a further covariate is no longer significant.
Multiple Testing Corrections
It may be possible to obtain a good test statistic by chance alone. Multiple testing corrections are designed to help ensure, if possible, this is not the case. You may optionally select one or more multiple testing corrections, including Bonferroni correction, false discovery rate, and single or full scan permutations tests.
