This is Penn State

Instructions

 

Predicting RNA-Binding Residues

RNABindRPlus predicts RNA-binding residues from protein sequences by combining the output from a Support Vector Machine (SVM) classifier with the output from a Homology-based method. Only the protein sequence is used as input by RNABindRPlus for generating predictions.

Protein sequences should be submitted in FASTA format; multiple sequences can be included in a single submission. You will receive your results via e-mail; expect a wait of approximately 10 minutes per sequence submitted. The rate-limiting step is the calculation of the PSSM (using PSIBLAST). To speed up the process of getting results from RNABindRPlus, we suggest that you submit not more than 30 proteins in a single submission. So, for example, if you would like to submit 100 proteins, break it up into smaller jobs of 20 - 30 proteins each.

The homology-based method (HomPRIP) that is used here exploits interface residue information from sequence homologs of the query protein(s). Users have the option of discarding highly similar homologs at a desired threshold (the default is set to 95% sequence identity) or utilizing all homologous sequences for making interface residue predictions. For rigorous comparison with other methods, the threshold can be reduced to ensure that the query protein and homologous sequences are not the same.

After trying your hand at RNABindRPlus, we welcome feedback to improve the program and its ease of use. For questions or suggestions pertaining to this website and RNABindRPlus, please contact: rnabindr_plus (at) iastate.edu

Format of Output

The output contains:

  • Sequence name
  • Amino-acid sequence
  • Prediction from HomPRIP
  • Predicted scores from HomPRIP
  • Prediction from SVM
  • Predicted scores from SVM
  • Prediction from RNABindRPlus
  • Predicted scores from RNABindRPlus
The result file is best viewed in a text editor, such as Notepad++ (in Windows) or Geany (in Unix). The contents of the result file are tab-separated and easy to read using a script. Contents can also be copied into a Microsoft Excel sheet and viewed.

How to Interpret the Output?

Your results are returned in a plain text format. A prediction of '1' indicates that the residue is predicted to be an interface residue and a prediction of '0' means that the residue is predicted to be a non-interface residue. The prediction score tells you the probability of a residue being an interface residue. If the probability is equal to or greater than 0.5, then the residue is predicted to be an interface residue. If the probability is less than 0.5, then the residue is predicted to be a non-interface residue.