Researchers have gathered data on a large number of proteins and stored information on their amino acid sequences in databases that are publically available. For some of these proteins researchers do not have information on the function of the protein. One method for predicting the function of a protein is to identify what class of molecules the protein may bind to.

The aim of the tool on this web site is to make predictions on which amino acid residues in a protein sequence have the potential to bind to RNA. A researcher can submit a protein sequence to the tool and get back information on which amino acid residues may bind to RNA molecules.

The tool is based on a mathematical model that integrates information about the amino residues in the protein sequence. The mathematical model has been developed by taking information from protein sequences where we know which amino acid residues bind to RNA. Using information about these known RNA binding amino acids we are able to develop a mathematical model which we can use to make predictions for sequences where we do not know which residues bind to RNA.

The tool can be used to make theoretical predictions about protein function without carrying out experiments in the laboratory, which are expensive in terms of time and money. If researchers have no clue about the function of a protein they have to perform a large number of experiments to narrow down the possibilities. However, if researchers have an indication as to what type of molecules a protein binds to they can perform far fewer experiments to confirm its function. This is how the tool presented here is useful. Researchers can make initial theoretical predictions, and then confirm protein function using more targeted laboratory experiments, that would not be possible if information about RNA binding was not known.