Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

From: A method for the prediction of GPCRs coupling specificity to G-proteins using refined profile Hidden Markov Models

Figure 2

Flow chart of the method. GPCR entries in UniProt of known, non-promiscuous coupling specificity to G-proteins summarized in [37] were used to extract intracellular regions sequences, based on membrane topology information of the UniProt annotation. ClustalX was used to generate multiple sequence alignments of intracellular regions from which low-entropy blocks were selected based on ClustalX row scores. For every begin and end row, within low-entropy regions, sub-alignments were extracted and profile Hidden Markov models (pHMMs) were built. The discriminative power of each pHMM was assayed, after an hmmsearch run against the training set. The most discriminative HMMs for each intracellular region were selected for each one of the three main coupling groups and appended in the refined library. E-value thresholds were then set for each pHMM included in the refined library. The reverse course is followed during a query against the library of refined pHMMs.

Back to article page