proRBR server allows user to predict Protein-RNA Binding Residues (RBRs) and non-Protein-RNA Binding Residues (non-RBRs) for a given protein sequence. The model achieved prediction accuracy of 86% and Area under the curve (AUC) of 0.93.
proRBR-Manual
- Either type or paste the input protein sequence in FASTA format or upload a FASTA file using the Browse button.
Note that the input sequence should be a protein sequence with standard amino acids. The input can be a single line sequence or a single-line description followed by line of sequence.
- Click on the submit button to run the prediction. The webserver uses the default RF-NC-triplet model for prediction, once the user submit the query sequence.
- Users can optionally select other prediction models such as RF-NC-quartet, SVMFBF, or KNN developed using quartet nucleotide composition feature and other optimized ML algorithms using the select prediction model button.
- On submitting the query sequence, the output of prediction result is displayed on the screen. Each nucleotide is assigned a prediction annotation (1/0) for protein-binding and non-binding nucleotides along with their raw probability scores. Optionally, user can download the prediction results by clicking on the Download here link.
- The Test Run button allows user to check an example (test case) input sequence and the corresponding prediction results.