Skip to main content


Fig. 3 | Biology Direct

Fig. 3

From: xHMMER3x2: Utilizing HMMER3’s speed and HMMER2’s sensitivity and specificity in the glocal alignment mode for improved large-scale protein domain annotation

Fig. 3

Average coverage per domain model and average True-positive rate of the Pfam domain models (release 27). In a, an average for each domain model is computed by taking the total coverage of detected positive hits divided by the total detected positive hits. The average of a full domain coverage is one. An overall average is taken over all 16295 HMMER3-build Pfam models for each false-positive rate (FPR). Across the FPR range, the average domain coverage of the Pfam domain models hovers about 0.88 as indicated by the horizontal dotted line. b gives the distribution of domain length of the Pfam library. There are a total of 1801 domains with length of 485 alignment positions or more (i.e., larger than 1.5 times the inter-quartile range above the median; see vertical line in b)

Back to article page