Skip to main content
. 2020 Mar 9;21:62. doi: 10.1186/s13059-020-01972-x

Table 2.

Area under the precision-recall curve (PR-AUC) for all algorithms applied to TKO datasets, using the combined non-human- and Chr10Rand-negative control guide RNAs

HCT116_1 HCT116_2 DLD1 HeLa RPE1 GBM Average rank
MAGeCK RRA 0.449 0.505 0.508 0.424 0.450 0.461 2
MAGeCK MLE 0.404 0.476 0.372 0.395 0.402 0.346 3.7
CRISPhieRmix 0.459 0.435 0.032 0.391 0.502 0.535 2.7
RSA 0.346 0.415 0.439 0.340 0.389 0.368 4.5
t test 0.431 0.535 0.57 0.383 0.474 0.528 2.2

Each column corresponds to a specific cancer type. The average rank is calculated by taking the rank (1 to 5) for each screen and taking the arithmetic mean. The highest PR-AUC for each dataset is highlighted in bold