Size-dependent Copy Number Analysis Scripts

Scripts used for the analysis in Pan-cancer copy number analysis identifies optimized size thresholds and co-occurrence models for individualized risk stratification. 

Meningioma

IDATtoSeg.R converts standard methylation Illumina IDAT files to segmentation files

CNV_cumulative_single uses sliding thresholds to call arm-level CNAs

CNV_cumulative_AUCs_single.R calculates univariate area under the curve for 5-year LFFR and OS for each arm-level CNA

CNV_bin_single.R calls focal CNAs at bins of a set size

BinAUC.R calculates univariate area under the curve for 5-year LFFR and OS for each focal CNA

FocalRegions.R generates plots and tables of focal ares of recurrent amplification/deletion

ModelBuilding.R and XGboost.R are for training and testing of meningioma size-dependent CNA models

Cooccurrences.R and CooccurrenceSurvplots.R are for looking at pairs of co-occurrent arm-level CNAs

CNAburden_clustering.R clusters arm-level CNAs and visualizes based on CNA burden

TCGA

CNV_cumulative_single_TCGA.R uses sliding thresholds to call arm-level CNAs

PanCan_sizedependent_loop.R calculates size-dependence of TCGA arm-level CNAs

PanCanSizedepdendentHeatmap.R generates descriptive heatmap of size-dependence across TCGA cancers

PanCan_cooccurrence_loop.R looks at pairs of co-occurrent arm-level CNAs