In this version of Smooth Descent four IBD calculators are implemented: the naive and weighted average (aka prediction) methods, as described in the preprint and the heuristic and hidden Markov model methods implemented in polyqtlR. These novel methods can be combined to estimate genotyping errors (with the best performance obtained with naive-hmm combination).
Additionally, we improved the implementation of MDSMap, our central linkage mapping calculator. Previous versions provided a sub-optimal map order, particularly in higher ploidies.