Search a protein database (KIND) through a HMM model

The search uses a protein query in FASTA format and runs it against the KIND - Karolinska Institutet Non-redundant Database using the position independent Smith-Waterman algorithm (swp). The search can be done using several standard substitution matrixes like BLOSUM62 or PAM250. The matching protein sequences are multiple aligned by a hierarchical method using the ClustalW software. A position-specific scoring matrix or HMM is then calculated and calibrate with the HMMER package. The HMM is searched against the KIND database. By multiple aligning the output and building a new HMM the user can choose to iterate several times in order to get a better result.

Search a HMM database (Scop, pfam)

The search compares a protein sequence against a database of HMM models like pfam and Scop. HMM searches perform position independent gap scoring using the Viterbi algorithm.