NEWS

ActiveDriverWGS 1.2.1 (2026-03-13)

Bug fixed: elements and sites with coordinates below 2 or beyond the chromosome end now return an informative error, preventing downstream failures in trinucleotide context calculations.
Bug fixed: all GenomicRanges operations (findOverlaps, setdiff, intersect, union) now use ignore.strand = TRUE, ensuring mutations, elements, and sites are always matched by genomic position regardless of strand.

Updated trinucleotide quantification in sequence and mutations. Trinucleotides on the inverse strand are now reverse complemented instead of complemented to match COSMIC signatures.
More accurate and stable numeric estimates of expected mutations are provided as output instead of the Poisson-sampled estimates provided previously.
The package now allows identification of genomic elements with significantly fewer mutations than expected, potentially reflecting negative selection of elements. This is enabled using an optional command line parameter detect_depleted_mutations. In the analysis depleted mutations, elements with enriched mutations are assigned non-significant P-values.

Bug fixed: datasets with very few unmutated elements (<10) previously failed to include these elements in results.
New option: mitochondrial mutations can be analyzed (chrM).

Bug fixed: a rare set of very large elements with depletion of mutations would sometimes appear as enriched in mutations.

Alternative reference genomes for human (hg19, hg38) and mouse (mm9, mm10) are now supported. Default option is hg19.
Full rewrite of ADWGS_rest for memory and speed efficiency.
Improved handling on indels on boundaries of elements is now default. Removed parameter element_bias that controlled this behaviour earlier.
README.rd and front page now include a simple example of running ActiveDriverWGS, and the vignette has been updated.
Site IDs need to match element IDs. Example databases with PTM sites have been updated.