As a first step, we build an HMM from every given PWM, and each HMM-promoter pair is assigned with a score reflecting the likelihood that the respective TFBS appears in the respective promoter region.
We expanded this family by iteratively searching a set of all proteins from completed prokaryotic genomes with the current version of the HMM, aligning the set of matching sequences, building an HMM from the alignment, and starting again with a new sequence search.
A recent method called FATHMM [ 39] also uses an MSA to identify conserved amino acid residues, but builds an HMM from the MSA and computes the differences in model probabilities between the wild-type and mutant residues to estimate the impact of the substitution.
HMMER3.1b (Eddy 2011) was used to build an HMM from the DUF1220 (PF06758) seed alignment stored in the PFAM database (Finn et al. 2014).
The popular tool HMMer can build a profile HMM from a family of aligned sequences (hmmbuild), search a profile HMM against a sequence database (hmmsearch), search a sequence against a profile HMM database (hmmpfam), and align a group of sequences against a profile HMM (hmmalign).
A preliminary protein family definition, based on an HMM constructed from a multiple sequence alignment of this putative ortholog set, marked a set of loci and created a phylogenetic profile, from which conserved gene neighborhoods and phylogenetically correlated protein families could be investigated.
Pfam-A HMM model was downloaded from the Pfam database [ 47].
To further explore the evolutionary affinities of the PafA family we prepared a multiple alignment and used an HMM derived from this alignment for an HHpred profile-profile comparison search against a library of HMMs derived from non-redundant PDB structures as seeds.
The alignment was used as input for the secondary structure prediction program JPRED, which combines the information from the frequencies of residues in the multiple alignment, a HMM derived from it and the PSI-BLAST Position-Specific Scoring Matrix (PSSM) to derive a final consensus predicted structure.
Kinannote uses an HMM derived from a manual alignment of the Dictyostelium kinome (Goldberg et al. 2006).
