The CAMPS system
The transmembrane domains of most integral membrane proteins consist of bundles of tightly packed alpha-helices.
Those transmembrane helices (TMH) interact specifically by electrostatic and Van der Waals forces.
The CAMPS system (Computational Analysis of the Membrane Protein Space)
is a resource specially tailored for classification and analysis of integral membrane proteins (3 or more TMH), providing extensive protein information and database integration.
CAMPS posses its own membrane protein organization based on single-linkage sequence clustering to generate structurally correlated groups (SC-clusters) and additional subclustering
satisfying orthology and 30% identity thresholds to produce ortholog homogeneous (OH-clusters) and modeling distance (MD-clusters) groups, respectively. As a result, a hierarchical system was built addressing the
three key levels of granularity in structural genomics: fold, function and modeling distance.
The CAMPER tool works as the protein classifier of the CAMPS system. CAMPER makes extensive use of a library of Hidden Markov Models constructed for every conserved TMH (TMH-HMMs) in
the SC-, OH- and MD-clusters. Then, every cluster is represented by a set of TMH-HMMs called signature which captures the specific sequence regularities of the transmembrane section of the cluster members.
The complete library of TMH-HMM is scanned on query sequences by the hmmpfam program of the HMMER suite, and subsequently parsed by CAMPER to detect the best profile signatures by evaluating the
global E-value, the average E-value per TMH-HMM and the percentage of TMH-HMM detected.
120 genomes (Archaebacteria and Eubacteria)
72% coverage of proteins with at least 3 TMHs
266 SC-clusters (fold level)