A CRISPR-drug perturbational map for identifying compounds to combine with commonly used chemotherapeutics.

We used t-SNE 28 and UMAP 29 to cluster the data, revealing an interesting trend, with the data primarily clustered by cell line, rather than by drug or outgroup status (Fig. 2a ; Supplementary Data Fig. 2a–d ; UMAP parameters were selected using an automated Monte Carlo approach, see Methods).

also in 723 other papers

We investigated this by a simple and interpretable orthogonal analysis, calculating the number of cell lines crossing a nominally significant RRA < 0.05 for each drug (Fig. 2b and Supplementary Data Fig. 2e ) and for each cell line (Fig. 2c and Supplementary Data Fig. 2f ).

also in 3,108 other papers