Code to reproduce simulation results, figures, and real data analysis results from the paper "Selective inference for hierarchical clustering" by Lucy L. Gao, Jacob Bien, and Daniela Witten.
naive_type1.R produces the results displayed in Figures 1-2.
Running our_type1.R with this.sim.id between 1-6 produces the results displayed in Figure 4.
Running cond_power.R with this.sim.id between 1-3 produces the results displayed in Figure 5.
Running our_type1_est.R with this.sim.id between 1-4 produces the results displayed in Figure S1.
Running power.R with this.sim.id between 1-9 produces the results displayed in Figures S2-S3.
Contains the results from running the code in the simulation-code folder as described above.
Produces Figures 1-6 in the paper and Figures S1-S3 in the supplement. All figures except Figures 3 and 6 depend on the simulation-results folder.
Contains the results from calling the code in the figures-code folder.
The code to run the real data analysis in Section 6.1 is in penguins.R. Instructions on how to download the data can be found here.
The code to run the real data analysis in Section 6.2 is in zheng.R. The T-cell data can be downloaded here, the B-cell data can be downloaded here and the monocyte data can be downloaded here. You'll need to put the three resulting folders in the raw directory inside real-data-code and rename the three folders to 'filtered_matrices_mex_memory', 'filtered_matrices_mex_bcell', and 'filtered_matrices_mex_mono'.