- Modifies the segmentation algorithm to prevent large segments from spanning CpG data gaps. The new algorithm will automatically create segment break points between two consecutive CpGs that exceed a maximum gap threshold. This reduces segment size inflation around regions with low coverage and/or absent phasing information.
- Adds a CLI option
--max-gap
to control the maximum gap threshold, default: 1000 bp.
- Fixed a panic caused by low sample sizes in the
compare
command. These will now be flagged asInsufficientData
.
- Adds a CLI option
--enable-haplotype-segmentation
that will run segmentation on the individual haplotypes (H1 and H2) and save those segments in the output BED file. For example, an unmethylated region on haplotype 2 will have the label "H2_Unmethylated" (or "H2_U" if condensed). - Adds a CLI option
--enable-nodata-segments
that will generate segments where haplotype data is missing. These will be flagged like "H1_NoData" (or "H1_ND" if condensed).
- Adds a new segmentation mode based on Circular Binary Segmentation. This mode generates a BED file of methylated, unmethylated, and ASM regions. The mode is executed with
methbat segment ...
, see README for more details.
- Improved error handling for all CLI interfaces
- Initial alpha release
- Includes initial version of HPRC background profiles