I am trying to use the meshSim function in the meshes package to calculate pairwise semantic similarity for vectors of MeSH terms, by their DUI. I am passing hsamd to the semData function, as suggested in the documentation:
meshSim(mesh_vec1, mesh_vec2, semData = hsamd, measure = "Wang")
But I am realizing that the default setup for the hsamd includes only the A branch of the MeSH tree (anatomical terms), and the MeSH terms in my dataset come from across multiple branches of the tree.
I have looked at the documentation for the meshdata function that is used to construct the hsamd data to see if it's possible to incorporate all of the branches of the MeSH tree, such as by modifying the code thusly (in this case to include branches A, B, and C:
meshdata("MeSH.Hsa.eg.db", category=c('A', 'B', 'C'), computeIC=FALSE, database="gendoo")
However, when I run this, it just seems to hang forever and I end up restarting my R session without it completing. The documentation for the meshdata function includes the info that "category" argument is described as "one of supported category," Does this really mean that this package can only calculate semantic similarity of terms in only one single branch of the tree?
For reference, here is the relevant session info:
R version 3.5.1 (2018-07-02) Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows >= 8 x64 (build 9200)
other attached packages:  MeSH.Hsa.eg.db_1.11.0 MeSHDbi_1.18.0 BiocGenerics_0.28.0 meshes_1.8.0
Any advice on this is much appreciated!