Analysis Tools 
Look beyond one’s own nose: Integrating results from (genomewide) association studies with bioinformatic information In the following screencast, we show, how results from association studies, whether they are from own GWAS or publicly available GWA results, can be embedded into the UCSC Genome Browser, which provides comprehensive functional genomics data. These results are then aligned with the implemented annotation tracks, including amongst others information on regulatory elements and interspecies conservation. Combining both statistical and biological perspectives can reveal an association which would have been missed by standard GWA approaches and can be used for the planning of functional studies. Since the integration of own custom data tracks (pvalues from GWAS or LD correlation) is not straightforward, we provide both a R function and a Microsoft Excel 2007 file, which do the necessary transformation. Microsoft Excel 2007 file "prepareUCSC.xlsm" Reference: Lamina C, Coassin S, Illig T, Kronenberg F: Look beyond one’s own nose: Combination of information from publicly available sources reveals an association of GATA4 polymorphisms with plasma TG. Atherosclerosis 2011 Dec;219(2):698703.
2016/03: Modified script version, which allows the creation of classic, flexible custom tracks with elements of variable size (without pvalues and correlation). Can be used to show deletions, repeat regions, probes etc. in the UCSC genome browser. Microsoft Excel 2007 file "prepareUCSC_flex_coord.xlsm"
Interaction terms are often included in regression models to test whether the impact of one variable on the outcome is modified by another variable. However, the interpretation of these models is often not clear, if two continuous variables are involved. To ease the interpretation of interaction effects between two continuous variables, we propose a graphical visualization approach which we implemented in functions in the statistical program R. They can be used on Generalized Linear Models (Linear and Logistic Models) as well as Cox Proportional Hazards models. The mutual modifying effect of both variables constituting the interaction effect can be shown by a twodimensional twisted surface (Figure 1), a colored contour plot (Figure 2) as well as interaction plots showing the effect estimator of one variable for varying values of the other variable. To apply the functions on your regression models, download the file “Plot_interaction_3D_2D.R” and include it into your R workspace with the “source”command. An example showing this visualization approach with possible interpretations can be found in the file “Plot_interaction_3D_2D_example.R”. Reference: Lamina C, Sturm G, Kollerits B, Kronenberg F: Visualizing interaction effects: A proposal for presentation and interpretation. J Clin Epidemiol. 2012 Aug;65(8):85562.
R function "Plot_interaction_3D_2D.R" R function "Plot_interaction_3D_2D_example.R" Figure legends: Figure 1  Interaction surface plot. The zaxis shows the predicted values of the outcome variable (vary) for varying values of variable 1 (var1) and variable 2 (var2), illustrating their interacting effect. Figure 2  Filled Contour plot. The predicted values of the outcome variable (vary) for varying values of variable 1 (var1) and variable 2 (var2) is displayed in different colors. The definition is given in the plot key, ranging from blue (low values of vary) to pink (high values of vary).
/media/analysis/Plot_interaction_3D_2D.R
