RNA folding with non-canonical basepairs and base-triplets.
The algorithm implemented here-in provides extended RNA secondary structure prediction. Each predicted nucleotide pairing is extended with an annotation describing which of three nucleotide edges is engaged in the pairing. In addition, each nucleotide may be engaged in more than one pairing.
The algorithm is described in
Hoener zu Siederdissen C, Bernhart SH, Stadler PF, Hofacker IL,
"A Folding Algorithm for Extended RNA Secondary Structures",
Bioinformatics (2011) 27 (13), i129-136
http://www.tbi.univie.ac.at/software/rnawolf/
Please note that "experimental" does mean experimental. We are mostly concerned with determining a good set of (heuristic) rules for run-time reduction currently. This version does include stacking and is able to fold sequences of a few hundred nucleotides in seconds.
Triplet calculations will come back with the next version (in a few days). The recursions require a number of changes to keep the runtimes down (as has been done for the extended loops without triplets).
We have recently split the Biohaskell libraries into smaller individual libraries. In addition, stacking, intermediate arrays, fusion and newtype-wrapping did require a number of changes. Please send a mail, if you encounter strange behaviour or bugs.
Changes in 0.4.0.0
secondary structure constraints (untested and undocumented)
Changes in 0.3.2.0
simpler training wrapper
added parallelism option for multi-core systems (reduce iteration time for the cost of a possible reduction in training efficiency; but should be worth it)
Changes in 0.3.1.0
fixed bugs introduced by bulgeinteriormulti-loops