Linking to scop

Structural Classification of Proteins

Warning

A new set of identifiers for all nodes in the SCOP hierarchy and corresponding parseable files will be introduced in the next release. If you are still referring to SCOP via classification page numbers (scop...html), despite their usage was deprecated since 1994, be aware they will disappear in the next release. What said below will still apply, and new options for linking/referring to any level of the SCOP hierarchy based on the new identifiers will be available from next release on.

Please don't hesitate to get in touch with us if you have any further questions about reliably linking to SCOP. Broken links or, worse, misleading ones, may drastically reduce the usefulness of information on the web.

Linking to scop

As documented in the help page since 1994, the filenames for the SCOP HTML files (classification page numbers of the form 1.002.044.001.002.021 that appear in dir.dom.scop.txt, col 4) are arbitrary and subject to frequent change. They can be used consistently for the analysis of a SCOP release, but they should never be used to identify (or link to) the same domain across releases. Please use the search engine instead, which has been available from the very beginning and provides a reliable way of dynamically linking to various levels of the hierarchy.

The best way of linking to SCOP depends on the logic behind it. Here is a brief summary.

If you are interested in the information related to a specific pdb entry, use:
http://your favourite scop server/search.cgi?pdb=xxxx
where 'xxxx' is the pdb identifier. Note that there can be more than one SCOP domain corresponding to a given pdb file. In this case, a list of links will be returned. Try 1dan, for example.
If you are interested in the information related to a specific protein domain, use:
http://your favourite scop server/search.cgi?sid=dxxxxxx
where 'dxxxxxx' is one of the SCOP domain identifiers, or sid, appearing in the first column of dir.dom.scop.txt. From year 2000 on (release 1.50) these identifiers are expected to be stable across releases, unless the domain definition itself changes or a pdb entry becomes obsolete. Even before, they rarely changed.
In both cases, for a pdb entry or a SCOP domain entry, you can specify the level of the hierarchy by qualifying your search using the tlev attribute. So:
http://your favourite scop server/search.cgi?pdb=xxxx&tlev=sf
will return a (list of) SCOP page(s) corresponding to the superfamily level in the hierarchy. Valid level identifiers are: cl (class), cf (fold), sf (superfamily), fa (family), and dm (domain).
Used with sid, the tlev attribute provides a link to intermediate nodes of the classification which is stable and does not vary across releases. For example:
http://your favourite scop server/search.cgi?sid=d2cro__&tlev=fa
points to the Phage repressors family page, independently of which SCOP release you are linking to, with the caveat that the classification itself could change. Nodes in the hierarchy can merge or split as more evidence becomes available from experimental data.

September 2000