Skip to main content

Table 2 Parameters used in tests of different protocols for the iterative generation of HMMs.

From: Application of protein structure alignments to iterated hidden Markov model protocols for structure prediction

Run Parameter Set (PS)

E-value Cutoff Progression (≤)

% Coverage Cutoff Progression (>)

R im Cutoff Progression (<)

Realign All Sequences To Model After Each Iteration?

1

10-25, 10-10, 10-5, 0.001, 0.01, 0.1

70, 70, 70, 50, 50, 50

10, 3, 3, 3, 3, 3

No

2

10-25, 10-10, 10-5, 0.001, 0.01, 0.1

70, 70, 70, 50, 50, 50

10, 3, 3, 3, 3, 3

Yes

3

10-6, 10-5, 10-4, 0.001, 0.01, 0.1

10 for all iterations

20 for all iterations

No

4

10-6, 10-5, 10-4, 0.001, 0.01, 0.1

10 for all iterations

20 for all iterations

Yes

  1. All runs proceeded through six iterations in order to build an alignment and corresponding profile HMM. Runs are named by their parameter set (PS). For each column describing cutoff values, the values are given in order of iteration cycle, from 1 to 6. In the case of E-value cutoffs, sequences were required to match the model with an E-value less than or equal to the given value. In the case of % coverage, sequences were required to align to a greater percentage of HMM match states than the given value. In the case of R im , (the ratio of inserts to matches, explained in the text) sequences had to score below the given value.
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us