SNPs&GO and SNPs&GO^3d have been optimized to predict if a given single point protein variation can be classified as disease associated or neutral. SNPs&GO requires in input:

• Protein Sequence: Protein Sequence: sequence-based SNPs&GO server requires the protein sequence that can be provided in three different ways: 1) raw format, 2) Swiss-Prot code; 3) uploading a text file containing the protein sequence. The server first checks the Protein Sequence box, if no sequence is provided it then checks if a text file has been uploaded. If both options have not been selected the server expects to receive in input the SwissProt code of the protein.
  
  
• GO terms: although this input is optional functional information significantly increases the accuracy of the method. If GO terms are not provided the expected accuracy of the method is comparable with PhD-SNP. GO terms used for the prediction are automatically retrieved only if the input sequence is a Swiss-Prot code. If not, user has to manually provide the GO terms and the server will automatically find all their parents
  
  
• Mutation: the server requires a list of comma separated mutations in the XPOSY format where X and Y are the wild-type and mutant residues respectively, and POS is the number of the mutated position in the sequence.
  
  
• All Methods:with this option the server will performs predictions using PhD-SNP and SNPs&GO methods. If not checked only the prediction from SNPs&GO will be returned. If not checked only prediction from SNPs&GO will be returned;

The structure based algorithm SNPs&GO^3d takes in input:

• Protein Structure: the protein structure can be provided using the PDB code or uploading your own PDB file.
  
  
• GO terms: although this input is optional functional information significantly increase the accuracy of the method. If GO terms are not provided the expected accuracy of the method is comparable with S3D-PROF. GO terms used for the prediction are automatically retrieved only if the input sequence is a Swiss-Prot code. If not, user has to manually provide the GO terms and the server will automatically find all their parents.
  
  
• Mutation:the server requires a list of comma separated mutations in the XPOSY format where X and Y are the wild-type and mutant residues respectively, and POS is the number of the mutated position in the sequence.
  
  
• All Methods: with this option the server will performs predictions using S3D-PROF and both SNPs&GO and SNPs&GO^3d methods. If not checked only prediction from SNPs&GO^3d will be returned.

The results of both algorithms can be sent to your e-mail address filling the appropriate box or displayed interactively if you do not enter any e-mail address.


Output
The output consists of a table listing the number of the mutated position in the protein sequence, the wild-type residue, the new residue and if the related mutation is predicted as disease-related (Disease or as neutral polymorphism Neutral). The RI value (Reliability Index) is evaluated from the output of the support vector machine O as