Instructions for use of RNABindR

PDF Version of these instructions

RNABindR provides two services:

1. Calculating RNA binding residues in a protein-RNA complex from the PDB.

2. Predicting RNA binding residues in a protein sequence that is not available in a protein-RNA complex in the PDB.

Calculating RNA binding residues

1. Enter the PDB id of a protein-RNA complex in the first text box. Click here for a list of available PDB ids.

2. If you are only interested in the interactions of a specific protein chain within the complex, you can enter the chain id in the second text box.

3. Enter the cutoff distance (in Angstroms) to be used to define RNA binding residues. By default, RNABindR uses a distance cutoff of 5 Angstroms.

4. Click the “Calculate” button.

For most protein-RNA complexes, results are returned within a few seconds. If you are analyzing a large complex, like a ribosome structure, the results may take a few minutes to appear.

The results consist of the sequence of the chain(s) in the complex with a “+” indicating RNA binding residues and a “-“ indicating non-RNA binding residues. A Jmol viewer is used to display the RNA binding residues on the complex structure. Protein is shown in blue spacefill, RNA is shown in green wireframe, and RNA binding residues within the protein are shown in red spacefill.

Predicting RNA binding residues

1. Enter a protein sequence in the large text box at the top of the page. The sequence can be in any format but FASTA format is preferred.

2. Click the “Predict” button.

RNABindR first searches for an exact match to the query sequence in the PDB. If an exact match is found in a protein-RNA complex, the prediction program is not run. Instead, the actual RNA binding residues are returned along with a Jmol viewer showing the complex.

Three sets of predictions are returned. The “optimal” prediction uses the settings for the classifier that maximized the correlation coefficient on our training set. The “high specificity” prediction predicts fewer RNA binding residues with higher confidence in the predictions. On the training set, 80% of the residues predicted to be RNA binding were actually RNA binding at the “high specificity” settings. The “high sensitivity” prediction identifies more of the actual RNA binding residues at the cost of more false positives. On the training set, 80% of actual RNA binding residues were detected at the “high sensitivity” settings. Specificity is the fraction of residues predicted to be RNA binding residues that are actually RNA binding residues. Sensitivity is the fraction of actual RNA binding residues that are predicted to be RNA binding residues.

Please see the Examples page for more information about running RNABindR.

PDF Version of these instructions