A brief introduction to HMMBUILD
The program HMMBUILD, belonging to the HMMER package, generates a profile HMM from a multiple sequence alignment. Profile HMMs are statistical models of the primary structure consensus of a sequence family. A profile HMM can typically be used in homology detection by comparing a query profile HMM against a DNA or protein sequence library.
For further details, see the HMMER homepage.

Availability in NPS@
HMMBUILD is available :

• At URL : https://npsa-prabi.ibcp.fr/NPSA/npsa_hmmbuild.html.
  The input multiple protein alignment must be in clustalw format.
  For hmmbuild, the standard character set for proteins is {A, C, D, E, F, G, H, I, K, L, M, N, P, Q, R, S, T, V, W, Y}.
• After a multiple alignment with CLUSTALW.

Parameters
First you have to choose the profile HMM construction options :

• the alignment mode :
  • Global mode : single global/global alignment.
  • Domain mode : multiple global alignments with respect to the profile HMM, and local with respect to the sequence.
  • Local mode : single local/local alignment.
  • Multi-hit local : multiple local alignments with respect to the profile HMM and local with respect to the sequence.
• the weighting mode
• the pbswitch option : above this threshold sequence number, hmmbuild will switch fronm GSC, BLOSUM, or Voronoi weights to PB weights which is faster.

NPS@ HMMBUILD output
The HMMBUILD output allows you to follow your analysis with a profile HMM search against a protein library thanks to HMMSEARCH.

References