prism_repurposing_primary The primary PRISM Repurposing dataset contains the results of pooled-cell line chemical-perturbation viability screens for 4,518 compounds screened against 578 or 562 cell lines. It was processed using the following steps: - Calculate the median fluorescence intensity (MFI) of each bead-replicate pair. - Remove outlier-pools. MFI values are log-transformed and centered to the median logMFI for each cell line on each detection plate. For each well on a plate, the median of the centered values is standardized (using median and MAD) against all other wells in the same position in the same screen. Data from wells with a standardized score greater than 5 or less than -5 is filtered. - Remove cell line-plates with low control separation. For each cell line-detection plate combination, the SSMD is calculated between the logMFI of positive and negative control treatments. Cell line-detection plates with an SSMD > -2 are removed. - Divide data from each well-detection plate by the median of control barcodes from the same well-detection plate. - Calculate log2-fold-change for data relative to negative-control wells from the same cell line on the same detection plate. - Positive-control and negative-control treatments are removed. - Run ComBat for each treatment condition (compound plate-well combination), considering log-viability values as probes and cell line pool-replicate combinations as batches. - Median-collapse each drug-dose-cell line combination **************** Dataset contents **************** ## Shared metadata: cell_line_info - table Cell line and control barcode metadata for rows of all PRISM Repurposing data matrices Columns: row_name - ID used to identify cell line or control barcode in data matrices depmap_id - ID to identify a cell line ccle_name - ID to identify a cell line primary_tissue - secondary_tissue - tertiary_tissue - passed_str_profiling - indicator for problems identified by STR profiling str_profiling_notes - summary of problems identified by STR profiling originally_assigned_depmap_id - depmap_ID assigned to cell line before identification of problems by STR profiling originally_assigned_ccle_name - ccle_name assigned to cell line before identification of problems by STR profiling pooling_info - table Pooling information for all profiled cell lines and control barcodes Columns: row_name - ID used to identify cell line or control barcode in data matrices depmap_id - ID to identify a cell line ccle_name - ID to identify a cell line screen_id - ID to identify a screen analyte_id - ID used to identify the luminex bead coupled to the cell line barcode detection_pool - ID used to identify the lysate detection pool of each cell line minipool_id - ID used to identify the mini pool that a cell line is thawed in before pooling pool_id - ID used to identify the pool that a cell line is assayed in passed_str_profiling - indicator for problems identified by STR profiling ## Replicate-level files: primary_replicate_treatment_info - Table Drug treatment metadata at the replicate level Columns: column_name - ID used to identify the column of replicate-level data matrices broad_id - ID used to identify drug-batch combinations name - name of drug dose - dose of drug(uM) perturbation_type - “positive_control” treatments are MG-132 or bortezomib and should have strong cytotoxic effects on all cell lines. “vehicle_control” treatments have been treated with DMSO. “experimental_treatment” includes all other treatments screen_id - ID to identify a screen detection_plate - ID of plates used to detect luminex barcodes for a single replicate of a compound plate compound_plate - ID of plate used to seed compounds well - plate-position of treatment detection_plate_scan_time - timestamp vendor_id - ID for drug vendor purity - purity of drug moa - Broad Repurposing Hub mechanism of action annotation target- Broad Repurposing Hub gene target annotation disease.area - Broad Repurposing Hub disease area annotation indication - Broad Repurposing Hub indication annotation smiles - simplified molecular-input line-entry system molecular identifier phase - Broad Repurposing Hub latest-clinical phase annotation primary_MFI - NumericalMatrix Median fluorescence intensities Columns: IDs identifying sample-replicate combinations. See column_name from primary_replicate_treatment_info Rows: IDs identifying cell lines or control barcodes. See row_name from cell_line_info primary_logfold_change - NumericalMatrix Logfold change values relative to DMSO, corrected for experimental confounders using ComBat Columns: IDs identifying experimental condition-replicate combinations. See column_name from primary_replicate_treatment_info Rows: IDs identifying cell lines. See row_name from cell_line_info ## Replicate-collapsed files: primary_replicate_collapsed_treatment_info - Table Drug treatment metadata at drug level, collapsed across replicates within a screen column_name - ID used to identify the column of replicate-collapsed data matrices broad_id - ID used to identify drug-batch combinations name - name of drug dose - dose of drug(uM) screen_id - ID to identify a screen moa - Broad Repurposing Hub mechanism of action annotation target- Broad Repurposing Hub gene target annotation disease.area - Broad Repurposing Hub disease area annotation indication - Broad Repurposing Hub indication annotation smiles - simplified molecular-input line-entry system molecular identifier phase - Broad Repurposing Hub latest-clinical phase annotation primary_replicate_collapsed_logfold_change - NumericalMatrix Replicate collapsed logfold change values relative to DMSO, corrected for experimental confounders using ComBat Columns: IDs identifying experimental conditions. See column name from primary_replicate_collapsed_treatment_info Rows: IDs identifying cell lines. See row_name from cell_line_info