compute_bio_phewas computes the PheWAS analysis on numeric biological test results
1 2 3 4 | compute_bio_phewas(bio, patients, direction = "SUP", bio_min_occur = 1,
zero_controls_add = 2, patients_min_concept = NULL, with_boot = FALSE,
perm = 1000, boot_all = FALSE, fdr_threshold = 0.05,
analysis_type = "poisson", progress = NULL, cores = 1)
|
bio |
bio data.frame from the 'get_data_from_num' function |
patients |
patients data.frame from the 'get_data_from_num' function |
direction |
one of c('SUP', 'INF') to perform the analysis on results above the normal range (SUP) or below the normal range(INF) |
bio_min_occur |
integer. minimum number of occurrences of the biological test result in the dataset to be analyzed |
zero_controls_add |
integer. number of simulated patients to add to the control group for a given if there is no occurrence of this concept in the control group. This trick allows to compute a pvalue in this situation. |
patients_min_concept |
integer. minimum number of unique concept for a patient to be included in the analysis |
with_boot |
logical. perform a bootstrap on the results |
perm |
integer. number of permutations for the bootstrap |
boot_all |
logical. perform a bootstrap on all the results or only on the results with FDR corrected p.value < fdr_threshold |
fdr_threshold |
maximum FDR corrected p.value to perform the bootstrap |
analysis_type |
one of c('poisson', 'binomial') for the type of analysis to perform |
progress |
progress bar object |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.