- Faster and more efficient index creation (libdivsufsort), requires about 6n main memory (for fasta files >2GB about 10n main memory). n is the size of the input fasta files. No additional secondary memory is needed (unlike for Skew)
- slightly smaller indices (about 0.625n - 0.75n less space). Old indices are still compatible with GenMap 1.3.0, but indices built with 1.3.0 are not compatible with older versions of GenMap.
- indexing will search for fasta files recursively (subfolders were not considered before)
- updated paper information (replaced bioRxiv with OUP Bioinformatics)
- BREAKING CHANGE! k-mers are searched on both strands by default. Can be turned off with
--no-reverse-complement - bedgraph output (
*.bg) replaces bed output (*.bed) (bed output is still supported with--bedbut deprecated and removed from the help string) - allow user to specify a filename with
--outputif only a single fasta file has been indexed (previously only the directory could be specified)
- truncate fasta identifiers after first space
- allow indexing fasta files with
*.fasfilename ending - runtime speedup when BED file is provided for computation on a subset of the input (
--selection) - reduced progress output on terminal when processing multiple fasta files
- compute mappability of selected regions using a bed file
- suppress 0 values in BED and WIG files
- change default algorithm for indexing to Skew
- BED output format fixed (end position was off by one, i.e. closed interval instead of half-closed interval)
--frequency-smallwould output unreadable ascii characters in--txtformat- some typo fixes
- faster computation of mappability for 3 and 4 errors
- included the reference to the paper (preprint)
- minor fixes in documentation and error messages
- preliminary version of GenMap released