oocyte data imputation

sim_find_fn

R Documentation

This function finds the number of false negatives when comparing two sets of recombinantion breakpoints

Description

This function finds the number of false negatives when comparing by finding the locations for which nothing intersects true recombination locations. When multiple predictions intersect a single truth, in the conservative approach, we consider all but one of the the intersections to be false negatives because we in essence aren't predicting a true recomb spot, or we're falsely saying there's not a recombination spot there. If there are no non-NA truths but some non-NA predictions, there are 0 false negatives. However, if there are some non-NA truths, but no non_NA predictions, there are the same number of false negatives as there are non-NA truths because the predictions never predicted breakpoints for those truths

Usage

sim_find_fn(
  truth_intersect_dt,
  truth_dt_nona,
  cons = FALSE,
  no_truths = FALSE,
  no_preds = FALSE
)

Arguments

`truth_intersect_dt`	a data table from foverlaps intersecting the truth and the predicted data with columns of `gam`, `Predicted_Start`, `Predicted_End`, `True_Start`, `True_End`
`truth_dt_nona`	a data table of just the truth data with no NAs in `start` or `stop` columns.
`cons`	a bool, default is FALSE; only applicable when multiple predictions intersect a single truth. if `cons` is TRUE, we take a conservative approach and consider all but one of these multiple intersection breakpoints as false negatives
`no_truths`	a bool, default is FALSE, if true, that means there were only NAs in the truth information and no real recombination break points
`no_preds`	a bool, default is FALSE, if true, that means there were only NAs in the prediction information and no predicted recombination break points