Out of curiosity, I tried running a default NT search with almost all boxes checked for
Puschel et al.’s (2025) mammaliaform matrix with unpartitioned EIW at
k = 25.79102 (the value given from setk). Before the search, I got the following error regarding the xlinks:
Some characters (e.g. 48, determining applicability of 49) are missing or ambiguous in outgroup taxon (0). This can make evaluations of homoplasy imprecise! If possible, score outgroup unambiguously. N.B.: If you have not defined any asymmetric transformation costs, you can select a different taxon as outgroup without changing costs.
My attempted NT searches with the XSS option in Sectorial Search gave me four internal errors in TBR calculations before aborting. Without XSS, I got six TBR errors. Without only standard IW, I get 14 errors. Only after turning off implied weighting does the NT search run as expected for a discrete character matrix of this size with or without xlinks (~40 min w/o parallelization).
I’m curious if others have run into similar issues attempting NT searches with both EIW and xlinks.
[Ideally I would try runs with
Grams et al. (2025) malacostracan matrix, but I’m having trouble reading the version with xlinks, due to apparent differences in formatting compared to Puschel
et al.]
Jacqueline S. Silviria
she/her/hers