The resource uses a standard CNV binning model with (in GRCh38) 3102 intervals of 1MB default size. Binned genome CNV values are available for all individual analyses (genome profiles), indicating the interval’s involvement in one or more CNV events, separated for gains and losses.
Additionally, for all all collations (e.g. profiles of a tumor type, or from the same publication), pre-computed interval CNV frequencies are available Alternatively, interval frequency maps can be generated on the fly from e.g. search results.
The binning does not apply to the variant data itself; i.e. genomic variants are stored and being queried with their original parameters.
Binned CNV mappings are generated through utility functions in the bycon package.
As a result, all samples have now 2x3102 values for gain and loss overlap fractions as well as - for samples with accessible original data - relative CNV intensities.
CNV frequencies are calculated for collations, or samples from search results, by counting the occurrences of non-zero status values for all given intervals in the selected samples, with an optional filter for a minimal fraction.
More information is made availanble in the Use Cases category.