associate
Association analyses for PolyGenius
Description
associate is an environment that bundles high-level helpers for running association analyses on PolyGeniusData objects. The public surface includes:
-
associate$regression()fits one-vs-all or many-vs-all association models. -
associate$compare()runs formal comparisons between models, effects, or groups. -
associate$mediation()fits mediation analyses for an exposure, mediator, and outcome. -
associate$mr()is the Mendelian-randomization entry point. -
associate$meta()pools compatible summary results across studies.
Summary-mode association helpers return rich PolyGeniusAssociations objects. These objects keep inferential result tables separate from plot-support artifacts and diagnostics, which allows downstream forest plots, heatmaps, survival curves, and meta-analysis to operate on one common result container.
Usage
associate
Format
An object of class associate (inherits from environment) of length 5.
Details
The regression family pages are grouped under the Associate reference section. They are summarized below:
| Family | Analysis focus | Documentation |
“lm”
|
Continuous outcomes via Gaussian linear regression. | LinearRegression |
“glm”
|
Binary outcomes via binomial logistic regression. | LogisticRegression |
“cox”
|
Right-censored survival outcomes. | CoxRegression |
“crr”
|
Competing-risk outcomes via Fine-Gray regression. | CompetingRiskRegression |
“km”
|
Grouped Kaplan-Meier / log-rank comparisons. | KaplanMeierRegression |
Mediation currently supports classical linear and logistic mediation workflows via the suggested {mediation} package. Mendelian randomization remains a declared entry point, but no estimator backend is implemented yet.