NEWS
triplex 1.2.0
NEW FEATURES
- It's now possible to set custom scoring and isogroup tables through
triplex.search interface.
- New triplex.score.table and triplex.group.table functions for getting
default scoring and isogroup tables to make the customization process
almost effortless.
- Dynamic algorithm optimization technique implemented as a computational
reduction based on minimal score option.
- Optimized usage of processor data cache - the computation was divided
into smaller pieces to prevent cache misses.
triplex 1.0.0
BUG FIXES
- Fixed P-value definition, Fixed P-value definition, now it depends on
sequence length. See vignette for details.
- Fixed splitting sequence into smaller computational pieces. The end
of larger sequences was not searched properly.
- Fixed handling of special IUPAC symbols (R, M, W, D, etc.). These are
now cut off as N and - symbols.
- Fixed scoring matrices for matches and geometrical compatibility. They
were combined with wrong types of triplexes. Among other things this
caused the output of the program to be slightly innaccurate for certain
types of triplexes.