SCEPTRE improves calibration and sensitivity in single-cell CRISPR screen analysis

Table 1 Statistical methods employed in single-cell CRISPR screen analysis. Parametric methods are non-robust to misspecified gene expression distributions, and classical nonparametric methods cannot adjust for confounders. Conditional resampling (implemented in this work as SCEPTRE) addresses both challenges

Method class	Example	Robust to expression model misspecification	Able to adjust for confounders
Parametric	Monocle [16]	No	Yes
Nonparametric	Virtual FACS [8]	Yes	No
Conditional resampling	SCEPTRE	Yes	Yes

ISSN: 1474-760X