This tool provides SIFT predictions for a given protein FASTA sequence by using aligned sequences provided by the user. Your query sequence is the sequence that you would like prediction on (i.e. the sequence in which you have introduced the amino acid substitutions).
The query sequence must be first among the aligned sequences you enter. Partial sequences should be flanked by Xes at the beginning and end of the sequence so that those positions are not considered gaps (and hence have liberal prediction).

User Input

Enter your protein query sequence along with related sequences (not necessarily aligned). These sequences are assumed to be functional, and is what SIFT will base its prediction on. The normal (functional) query sequence should appear first.

Name of file containing protein query sequences (fasta format).


Paste in your protein query sequences (fasta format).

Enter the substitutions of interest [format]:


Upload a file containing substitutions of interest [format]:


Remove sequences more than percent identical to query
This cleans out any polymorphic alleles (sequences in the database that are nearly identical to your protein but already containing the substitutions of interest). The contaminating sequences may cause the substitution to be predicted as tolerated.

Page last modified: May 2009