toxbootHitParamCI: Performs Hit Call and Model Winner Selection on Bootstrap...
In ericwatt/toxboot: Bootstrap Methods for 'ToxCast' High Throughput Screening Data

Description Usage Arguments Details Value

toxbootHitParamCI takes input dose response data and returns a matrix of sampled dose response values.

1	toxbootHitParamCI(dat_db, dat_pipe)

`dat_db`	data.table, a data.table resulting from a toxboot query minimally with columns m4id, max_med, hill_ga, hill_gw, hill_tp, hill_aic, gnls_ga, gnls_gw, gnls_tp, gnls_la, gnls_lw, gnls_aic, cnst_aic
`dat_pipe`	data.table. Minimally this data.table must have a column for m4id and coff (cutoff value, used in determining the hitcall). Often use the resulting data.table from `tcplLoadData(lvl = 5L)`.

Based upon mc5 processing in tcpl.

This function will first choose a winning model for each bootstrap replicate, based on that row's values for cnst_aic, hill_aic, gnls_aic.

Second, if the winning model is hill or gnls, the max_med is greater than the cutoff, and the winning model modl_tp parameter is greater than the cutoff, the hitcall for that row is 1 (boot_hitc == 1).

Finally, if boot_hitc == 1, values for modl_ga, modl_gw, and modl_tp are set to the corresponding values in the winning model. Else, these parameters are set to NA.

dat_boot, a data.table equivalent to dat_db with added columns coff (from dat_pipe), maic (winning model aic), modl (winning model), boot_hitc, modl_ga, modl_gw, modl_la, modl_lw, and modl_tp

ericwatt/toxboot documentation built on May 16, 2019, 8:41 a.m.