Fig. 2
From: Organizing the bacterial annotation space with amino acid sequence embeddings
A Distribution of the lengths of the 3-mer vectors in the Bacillus carbohydrate metabolism Protvec model. The shaded region corresponds to 3-mer vectors with a length greater than 16. B Comparison of the Bacillus carbohydrate metabolism Protvec model with the BLOSUM62 matrix. The number of occurrences (count) of each amino acid in 3-mer vectors with a length greater than 16 is compared with the value of each amino acid on the diagonal of the BLOSUM62 matrix