search

uniprot / enzymeportal

The EBI Enzyme Portal
http://www.ebi.ac.uk/enzymeportal/
Apache License 2.0
11 stars 4 forks source link

Improve Search Filters and Enzyme Summaries #200

Open computingfacts opened 10 years ago

computingfacts commented 10 years ago

Currently , this is the internal flow of the enzyme portal.

  1. A user searches for an item (keyword, SEQ, compound).
  2. User request is handled by EBI search and resolved to UniProt IdPrefixes.
  3. enzyme portal makes request to various servers (Intenz, UniProt etc) using the IdPrefixes to build enzyme summaries.
  4. The enzyme summaries are cached and search results are filtered based on all the available (loaded) summaries).

However, the approach is slow especially as the number of Idprefixes increases. time required to build summaries depends on the number of Id Prefixes.

To resolve this issues, we resorted to apply a different approach.

Here is a new approach we have considered.

  1. A user searches for an item (keyword, SEQ, compound).
  2. User request is handled by EBI search and resolved to UniProt accessions of which we group them as orthologues ( IdPrefixes ).
  3. we keep a List of the idPrefixes returned per request. (note: paginate the list)
  4. build the enzyme summaries from only a portion (e.g 10) of the inPrefixes.
  5. retrieve the remaining idPrefixes as the user clicks next (pagination) or filters for an item.
  6. cache the IdPrefixes List (application scope)
  7. Search results are filtered based on the IdPrefixes List.