Vast

You can find the server here
You can start your search directly here

1.Operation

• 1.1. User interface, on-line help, choice of parameters and databases

  In order to start you have to type in your name and email just for case the request fail and password to view the results further on. You have to provide information, if your protein contains multiple chains and decide if you want to search in entire PDB or only subset of PDB. Reading of basic information about how to use VAST takes about 15-20 minutes - recommended

• 1.2.Interactive versus E-mail results or notification

  interactive upload only, you have to store your password (at least 6 letters) and search ID, for latter browsing. You have to have new versions of browser to upload the file, size of the file is limited to 1MB. Results also only interactive. The upload can become difficult, because VAST can not handle all pdb files and some editing of pdb file can be necessary (especially deleting all items except from "ATOM" lines and reshuffling of columns)!

• 1.3. Documentation, access to literature references

  on the main page there is a hyperlink to paper in Curr. Opinion in Struct. Biol., but you can get it directly only if you have account on Medline (otherwise you have to sing to Medline) and only HTML format, there is also reference to paper in Proteins on the main page, hyperlink only to abstract.

• 1.4. Confidentality

  Coordinates of query structure and result are kept in directories on their server. They can be read by members of NCBI group involved in the VAST search project and by System administrator. Vast search is not secure service. The user ID and password are not encrypted.

• 1.5. Elapsed time between submission and results

  It depends on the length of the protein - from 5 minutes to 1 hour

• 1.6. How long are the results kept on the server ?

  For a week

2. Presentation of results

• 2.1. What information is reported for the hits?

  very transparent, contains a lot of options ( redundancy level of the database, which parameter to use for sorting...). The default setting contains RMSD, number of matched residues, sequence identity between matched structures and short description of matched structure. One can choose which criterion will be used to rank the results.

• 2.2. Measure(s) for goodness-of-fit and statistical significance?

  P-value.

• 2.3. Structure-based sequence alignment?

  yes, as a window under structure superposition, possibility to save it in many different formats

• 2.4. PDB files with superimposed structures?

  yes, interface with Cn3D ( up to five superimposed structures) or with Mage (only one structure)

• 2.5. Interface with visualization software?

  Cn3D, Mage and PDB visualization program (link to download program is from the result page only to Cn3D. Link to Rasmol and Mage is possible (with some endeavour) to find on main page

• 2.6. Hyperlinks to the other web-based resources (e.g., PDB)

  for each found neighbor you can get to MMDB structure summary, it contains link to PDB summary of the structure. It contains also links to Entrez and NCBI homepage

3. Implementation issues

• 3.1. Are there parameters the user can set to optimise the results?
  No.

• 3.2. Is there a choice of database? How much of the structural universe do they cover?

  you can choose between PDB and non-redundant subset of PDB (non-redundant subset based on Blast p-values; the level of redundancy is set to treshold p-value 10e ^-40)

• 3.3. How often are the databases updated?

  No information, but probably frequency of PDB update (once a week)

Comments

More about us