How genome/sciclone works
Project type and purpose
• This is an R/Bioconductor package (sciClone) for inferring the subclonal architecture of tumors from Variant Allele Frequency (VAF) and copy-number data. It provides core clustering algorithms, object definitions, I/O routines, and plotting functions.
File structure
• Top level: DESCRIPTION, NAMESPACE, README.md, LICENSE (R package metadata)
• R/: implementation code
• clustering.R – core clustering algorithms
• object.R – S4 class (“scObject”) and data‐model definitions
• plots.R – wrappers for 1D/2D/3D plotting (ggplot2, rgl, etc.)
• sciClone.R – user‐facing functions (sciClone(), writeCluster*, sc.plot*)
• zzz.R – package load hooks
• man/: documentation pages matching each exported function/class
• debian/: packaging scripts for Debian/Ubuntu
• README.md: high-level usage, dependencies, examples
README for additional insights
• Inputs: VAF tables, copy-number calls, optional region filters
• Outputs: cluster summary tables, PDF plots (1D/2D), animated GIFs (3D)
• Dependencies: IRanges, rgl, ggplot2, RColorBrewer, methods, NORMT3, etc.
• User interface: R functions installed via devtools or R CMD INSTALL
Elements
• Clear labels: “Input Reader”, “scObject Model”, “Clustering Engine”, “Plotting Module”, “Output Writer”
Detailed yet focused breakdown
• “sciClone()” as the central orchestrator
• Dependency cloud around the plotting layer and clustering layer
• Call out the S4 class as the “in-memory” data carrier