How to use PROFbval

Input to the server:

Users can submit a protein amino acid sequence or a sequence alignment. If the user uses sequence as an input PROFbval runs psiBLAST and creates a profile based on the alignment. If the user chooses to use multiple sequence alignment as an input he/she has to specify the format of the alignment.

A successful submission will result the following page:

Output from the server:

There are two output formats:

• Per residue prediction: a list of all the residues and their corresponding prediction values are displayed. The following values are shown:
• number - The number of the residue in the sequence
• residue - the residue type
• accessibility – solvent accessibility prediction by PROFacc - ‘e’ marks exposed and ‘b’ marks buried. A residue that is predicted to be ≥16% exposed to the surface is defined to be exposed.
• Bnorm – predicted normalized B-values.
• 2 state prediction – predicts whether a residue is flexible (F) or rigid (-). The prediction could be done in the strict mode (S) or the non-strict mode (NS).
• Reliability index (RI) – the reliability index reflects the strength of the prediction (values 0-9). High values indicate strong predictions.
• We marked in blue all residues that are predicted to be extremely rigid & exposed as these residues might have functional role. For instance. Here, residue D10 in RNase HI is in the active site of the enzyme. However, E6 does not have a known functional role.

2) Graphical format – The predicted normalized B-values can also be displayed in a graphical format. The graph output can be smoothed in different windows. For instance, these are the results with window size 3 for RNase HI. Active sites are marked by arrows.