ProValidator:

Help

Example:

Step 1: Click "New session" to generate a new session number. Please, annotate this number if you plan to continue the session in another moment.




Step 2: Select a initial set of sequences that form part of the group you want to define.

Example. Suppose you want to create a profile for IclR transcriptional regulators family: Search in Swiss-Prot/TrEMBL (www.expasy.org) sequences belonging to this group.



Then, put the sequences in FASTA format in 'Sequences' window.

Select a 'similarity threshold', a 'black list' (i.e. words that mustn't appear in the sequence's header for being choosen), and a 'keyword' (i.e. a word that must appear in the sequence's header for being choosen. Optional) and click 'Clusterize'.



After a few minutes, the result:



Then, you have to align the final set of sequences that will be the seed of the profile. For that purpose, we recommend align.genome.jp, based on CLUSTALW algorithm.


Step 3: Once the set of clusterized sequences is aligned, you have to select a conserved region in order to construct a profile with this region.

For that, choose the limits of the alignment.



There is 4 alignments modes:
  • Global alignment mode: The alignment must start at the beginning of the profile and at the beginning of the sequence and must end at the end of the profile and at the end of the sequence.
  • Domain global alignment mode: The alignment must include the entire profile; it may start and end at any position within the sequence.
  • Semi-global alignment mode: The alignment must start either at the beginning of the profile or at the beginning of the sequence and must end either at the end of the profile or at the end of the sequence.
  • Local alignment mode: The alignment is possible anywhere.



    Calibrate the profile consists on run profile against a shuffled database. Then, calculate score-normalization parameters in order to fix threshold score at 8.5.
    This process can take a few minutes, depending on the length of the profile.