Fig. 1.

The CES data integration pipeline to improve identification of cancer essential genes based on functional genetic screens and molecular feature data. Genome-wide shRNA and CRISPR-Cas9 based essentiality scores as well as molecular profiles for each cell line were obtained from public databases and literature. For a gene in a given cell line, a feature vector was constructed including CRISPR-based essentiality scores, shRNA-based essentiality scores, as well as mutation count, RPKM from RNA-seq, mRNA expression from microarray, and copy number variation. The aim of CES is to provide a data integration model to improve the consensus estimation of essential genes in cancer.