Skip to main content
. Author manuscript; available in PMC: 2020 Jan 1.
Published in final edited form as: Pac Symp Biocomput. 2020;25:463–474.

Fig. 4:

Fig. 4:

(a) Pairwise cosine similarity (ranging from 1.00 to −0.23) between embeddings for each theme, clustered by hierarchical clustering. (b) Precision at various recall levels for drug-disease score predicted by each theme.