Homepage

Belozersky Institute

GeneBee

Russian EMBnet Node

HELP ON SCREENING PROTEIN SEQUENCE AGAINST PROSITE DATABANK

REFERENCES:


ALGORITHM

One of the methods to select the functionally important protein sequence fragment is to compare the sequence with PROSITE databank, that collects all patterns (rules of the sequence fragment construction) for biochemically important protein fragments along with it's description.

The given program is not looking only at exact correspondence of a fragment to pattern, but also it selects those fragments that meet only part of patterns conditions: it could be "mistakes" in a found correspondence of a fragment to the pattern.

The method of screening is described in Biochemistry (Moscow) (Brodsky L. et al., v.60, �8, pp. 1221 - 1229, 1995). It based on idea that existence of the significantly large number of letter's pairs in the selected sequence with the same distance between pair members is as in a pattern. The number have to be large in comparison with corresponding numbers in random sequence. By the method a rather weak correspondence between a fragment and the PROSITE pattern could be selected, including selection of fragments with lengths that differ from pattern length.



QUERY SEQUENCE: (should be in one-letter code format):

Example:
        MNIFEMLRIDEGLRLKIYKDTEGYYTIGIGHLLTKSPSLNAA
        KSELDKAIGRNTNGVITKDEAEKLFNQDVDAAVRGILRN AKLKPVYDSL DAVRRAAL
        INMVFQMGETGVAGFTNSLRMLQQKRWDEAAVNLAKSRWYNQTPNRAKRVITTFRTGTWDA
        YK


PARAMETERS:

In the dialog window you should set the following parameters of the program: