PRECOG (Predicting coupling probabilities of G-protein coupled receptors) lets you predict coupling probabilites of Class A GPCRs to 11 G-proteins based on input sequence information (either WT or mutated). It is built by training logistic-regression algorithm on sequence- and structure-based features, derived from a set of Class A GPCRs, whose binding affinities were determined using the Shedding Assay technique.
You can run it by simply pasting/uploading either lists of protein accessions (UniProt/Gene symbols), sequence(s) (in FASTA format) or mutations, by simply appending the positional information (e.g. A623I), separated by a forward slash (/), to the protein ID.
Hang tight while your search is running...
Precog was developed in the Russell lab at Heidelberg University, 2019