qNABpredict - quick Nucleic Acids Binding content predictions

[The server is currently undergoing maintenance. We apologize for any inconvenience this may cause to our users.]

Help | Materials | Standalone code | References | Acknowledgments | Disclaimer | Biomine

qNABpredict webserver provides fast predictions of a fraction/content of nucleic acids-binding amino acids in protein sequences. These predictions can be used to identify nucleic acid binding proteins and to quantify size of the corresponding binding interface. Empirical tests show that predictions using the taxonomy specific models are on average more accurate than using the generic model (proteins of unknown origin).

Please follow steps below to make the predictions:

1. Select a predictor based on sequence information:

   

Select the appropriate kingdom of the protein, if known:


OR

Select, if source of the protein is not known


           

2. Upload a file with protein sequences, or paste them into text area

Please enter FASTA formatted protein sequence(s) (see Help section for details of input format).

Each protein length should have minimum 20 and maximum 30,000 residues. For batch file submission, number of sequences in each batch file should not exceed 2000. For larger submission please contact the authors.

3. Provide your email address (optional)

Please enter your email address in the following text area. A link to results of assessment will be sent to your email address once they are ready. The results will be also available in the browser window.

4. Predict

Click the Run button to launch qNABpredict.

Help

    The format of each input protein is as follows:
  • Line 1: >protein ID
  • Line 2: protein sequence (one-letter amino acid encoding)

The protein sequence from Line 1 to Line 2 should be formatted in the FASTA format.

Materials

    Explanation of DESIGN and TEST datasets
  • Line 1: >protein ID
  • Line 2: protein sequence (1-letter amino acid encoding)
  • Line 3: nucleic acid binding annotations where 1 denotes binding residues and 0 denotes non-binding residues

Standalone code

    qNABpredict is also available as standalone software (python script) that can be run on the end user's hardware. We distribute code as a convinient container that includes all necessery applications, scripts and data. The container should be run in a virtual machine setting using the free and open source VirtualBox application. Please follow the following steps to download and install the standalone version:
  • Step 1. Download and install the VirtualBox application that is available at https://www.virtualbox.org/. VirtualBox runs on Windows, Linux, macOS, and Solaris and supports a comprehensive collection of guest operating systems.
  • Step 2. Download the distribution of qNABpredict software at this link.
  • Step 3. Import the qNABpredict.ova file in the VirtualBox
  • Step 4. Run the virtual machine using the "Start" button
  • Step 5. Login with username "biomine" and password "biomine"
  • Step 6. Go to directory ~/qNABpredict
  • Step 7. Execute ./qNABpredict

References

Upon the usage the users are requested to use the following citation:

Acknowledgments

We acknowledge with thanks the following software used as a part of this server:

  • ASAquick- Prediction of protein accessible surface area