Skip to main content

Table 2 Memory for neighborhood storage for different alphabets with adapted neighborhood lengths

From: Optimal neighborhood indexing for protein similarity search

indexed neighborhood alphabet bits per character (α) neighborhoods length (L) total per index line (2αL) relative gain compared to Σ20 (1 – 2αL/ 110)  
Σ20 5 11 110 0% Memory
Σ16 4 12 96 13%  
Σ8 3 14 84 24%  
Σ4 2 19 76 31%  
Σ2 1 32 64 42%  
  1. for neighborhood storage is computed with respect to sensitivity/selectivity trade-offs shown on Figure 7.