OncoCRISPRDB: an integrated platform for CRISPR screening data analysis in cancer research

Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) screening has become an essential tool for systematically deciphering gene functions in cancer. However, the multidimensional data generated by this approach pose major challenges for data integration and in-depth analysis. The field lacks a platform that can uniformly integrate CRISPR screening resources while also providing customizable analytical workflows. To address this gap, we developed OncoCRISPRDB to enable the systematic integration of CRISPR screening data and support multiscale biological interpretation. To achieve this goal, we constructed OncoCRISPRDB with three core modules: (1) Database: we systematically integrated CRISPR datasets from the GEO database together with data mined from the literature; (2) Algorithm integration: we incorporated advanced algorithms, including Model-based Analysis of Genome-wide CRISPR/Cas9 Knockout (MAGeCK), the Negative Beta-binomial T-test (NBBT-test), and Screening Bayesian Evaluation and Analysis Method (ScreenBEAM), while also supporting pathway enrichment, gene dependency scoring based on differential causal effects, and phenotype similarity network analysis; and (3) Interactive visualization engine: we provide comprehensive visualization capabilities for gene screening under specific phenotypes, upstream or downstream regulatory networks, and cross-dataset comparisons. The OncoCRISPRDB platform currently contains data from 30 cancer types, 82 cell lines, 2 unique tissues, 36 drug-related phenotypes, and 28 phenotypes associated with tumor growth or metastasis. By integrating large-scale CRISPR screening data with multiple analytical methods, OncoCRISPRDB provides visualization tools for exploring gene function, pathway effects, and gene-scoring profiles, thereby enabling the systematic investigation of the functional scores and similarity patterns of different genes and pathways under specific phenotypes.