Description Usage Arguments Details Value Author(s) Examples

View source: R/sim_functions.R

Generate samples from the test statistic under the null distribution - here we take the average rates of clonal exclusivity across trees, and also the histogram for each patient over all pairs with the values # clon. excl./#trees.

1 2 | ```
build_null_test_statistic(avg_rates_m, list_of_clon_excl_frac_trees_all_pats,
num_pat_pair, num_pairs_sim, beta_distortion = 1000)
``` |

`avg_rates_m` |
The average rates of clonal exclusivity to be sampled from. |

`list_of_clon_excl_frac_trees_all_pats` |
The list of two lists.
The first one contains a list entry for each patient containing the
vector with the values of the information from each pair
in a patient of how often it was mutated across trees. The second list
entry is a list with an entry for each patient that is a vector with the
values of in how many trees the pair was clonally exclusive. The patient
ordering in the lists has to be the same as in |

`num_pat_pair` |
The number of patients the simulated pairs are mutated in. |

`num_pairs_sim` |
The number of simulated gene/pathway pairs to be generated. |

`beta_distortion` |
The value |

This function simulates gene pairs for the likelihood ratio test to
generate values from the test statistic under
the null. It draws the average rates of clonal exclusivity from the
ones provided by the user. That is,
the average rates of clonal exclusivity have to be computed first for
each patient. The number of patients the
simulated pairs are mutated in can be specified with `num_pat_pair`

.
This function can be used to build the
ecdf of the test statistic under the null hypothesis (see Examples).
The patients in which the simulated
pairs are mutated in are randomly selected proportional to the number
of pairs in a patient.

The return value is a tibble with the columns 'test_statistic', 'mle_delta', and num_pat_pair columns with the respective rates that were drawn for each of the patients, num_pat_pair columns with the respective number of mutated times across trees, and num_pat_pair columns with the respective number of times of being clonally exclusive across trees, and num_pat_pair columns with the rate that was distorted by the beta distribution. The 'test_statistic' is the test statistic of the likelihood ratio test. The 'mle_delta' is the maximum likelihood estimate of the delta for the elevated clonal exclusivity rate in the alternative model of the likelihood ratio test.

Ariane L. Moore

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 | ```
avg_rates_m=c(0.4, 0.3)
list_of_clon_excl_frac_trees_all_pats <- list(list(c(5, 4, 5), c(5, 4)),
list(c(4, 4, 3), c(3, 2)))
sim_pairs <- build_null_test_statistic(avg_rates_m,
list_of_clon_excl_frac_trees_all_pats, 2, 100,
beta_distortion=100)
ecdf_test_stat <-
ecdf(as.numeric(as.character(sim_pairs$test_statistic)))
plot(ecdf_test_stat,
main="ECDF of the test statistic when num_pat_pair=2")
# assume the observed test statistic t=6.0,
# compute a p-value given the ecdf of
# the test statistic ecdf(T) from the null distribution
# p_value=P(T>t | H_0 true)=1-ecdf(t) ## (upper-tailed test)
p_value <- 1-ecdf_test_stat(6.0)
``` |

Embedding an R snippet on your website

Add the following code to your website.

For more information on customizing the embed code, read Embedding Snippets.