When drug treatments bias genetic studies: data files
datasetposted on 08.08.2019 by Floriaan Schmidt, Hiddo J. Lambers Heerspink, Petra Denig, Chris Finan, Rolf H.H. Groenwold
Datasets usually provide raw data for analysis. This raw data often comes in spreadsheet form, but can be any collection of data, on which analysis can be performed.
Numerical data underlying the figures of PLOS ONE paper: "When drug treatments bias genetic studies: mediation and interaction".
The files "Data_Figure3-4" pertain to simulation results described in the paper and appendix.
The file "Data_Figure5" contains aggregated level data on the proportion and type of diabetes medication prescribed over a 3 year period. Data were obtained from a Dutch diabetes cohort 'GAINTT': https://www.giantt.nl/nl/
The file "Data_Figure6" contains mean difference and p-values of 100+ genetic variants and their association with longitudinal (over a 3 year period) HbA1c levels. Results are stratified for treatment modelling strategy.
These data have been shared inline with the journal's open access policy.