PatternQuery:Explorer

From WebChem Wiki
Jump to: navigation, search

When mining the whole Protein Data Bank for 3D structural patterns it is always better to look before you leap. This is one of the purposes of the PatternQuery Explorer application. This JavaScript/HTML is tailored for:

  1. Querying user submitted data (up to 100 structures).
  2. Query smaller subsets of the Protein Data Bank
  3. Tweaking queries before searching through the PDB database.

The last point would be probably of the most interest for the wide-range of users. In the following visual you will learn how to work with PatternQuery Explorer.

If the service is not working the way it should, please check your browser compliances in the technical details section.

Launch PatternQuery Explorer app

PatternQuery-Explorer1.png PatternQuery-Explorer1b.png
Follow either Explorer link or Query Playground tab at the Synopsis page, in order to launch the application. Create a new session or load a previous one.

Get started with PatternQuery Explorer

PatternQuery-Explorer2.png PatternQuery-Explorer3.png
The UI of the PatternQuery Explorer consists of several elements. At the top of the page you can find a link and a date until when all the results will be stored. Start with loading your data... Upload your own structures in one of the numerous supported formats (*.pdb, *.cif, *.pqr, *.mol, *.mol2). It is possible to upload multiple files at the same time or use a single ZIP file (*.zip) containing multiple files inside it. Similarly you can load entries from the Protein Data Bank.
PatternQuery-Explorer4.png PatternQuery-Explorer5.png
You can query a subset of Protein Data Bank based on numerous metadata. At first select a property you would like to filter by, then select a comparison method and finally fill in the demanded value. Please carefully read the hints and examples below each selected property. Different properties can have different means of providing input values. Note that you can also include logical expressions. For example: S & !N in Atom property means that just structures containing a sulfur atom and no nitrogen atoms will be considered. Multiple filters can be applied at the same time. Once you are happy with the specified filters, you can view a subset of the PDB. Either the First 15 entries based on their id or Random 15 entries.
PatternQuery-Explorer6.png PatternQuery-Explorer7.png
This list highlights a fraction of the PDB structures satisfying your criteria. You can also check, how many structures these conditions apply to. Browse structures and tick them to add to the dataset you would like to query. Do not forget to + Add Selected when you are happy with your selection. You can periodically fetch random structures satisfying your criteria and select those you would like to add to your dataset as well as browse through all the list of entries satisfying your criteria. After loading structures, browse the list of loaded entries in order to inspect further details at PDB.org or see their validation reports in ValidatorDB. You can easily modify the list as well. Either by further adding new entries or deleting the present ones. Also note the Log panel at the bottom of the web page, since it contains valuable information from the process of loading structures.

Executing Query & interpreting results

PatternQuery-Explorer8.png PatternQuery-Explorer9.png
When you have loaded at least a single structure, you can immediately start typing queries in the query box at the top of the page. If you are unsure on how to create query, first browse the principles behind the PatternQuery language and learn how to build a query. Case studies and language reference might come handy as well. After the query is executed, which usually takes no more than a few seconds, you can inspect them in the browser or download them for further processing. You can come back to this session later using it's web address, the results are kept for at least a month.