9
$\begingroup$

The completeness of a module can easily be checked by looking at the Definition entry associated with the module. For example, in module M00010, it is given as:

Definition  K01647 (K01681,K01682) (K00031,K00030)

which can be translated to:

K01647 AND (K01681 OR K01682) AND (K00031 OR K00030)

If this expression evaluates to TRUE, the module is considered complete.

I am wondering whether analogous information exists for a single reaction. For example, when looking at reaction R00352, one finds the following information about its Orthology:

enter image description here

But I can't easily find the logical relation that correspond to the KOs for this reaction. It could be:

K01648 AND K15230 AND K15231 

or:

K01648 OR K15230 OR K15231

or:

K01648 OR (K15230 AND K15231)

and so on.

In the above example, I know that the correct expression is:

K01648 OR (K15230 AND K15231)

i.e. one either needs either K01648 or both of the other two subunits.

Can this information be retrieved from KEGG for each reaction and if so, how?

$\endgroup$
2
  • 1
    $\begingroup$ Have you tried to asked KEGG directly via kegg.jp/feedback ? If anything, that might give them the information that there's something missing which people need. $\endgroup$
    – BaCh
    Commented May 30, 2017 at 21:17
  • $\begingroup$ You might need to look into the FTP dump; I looked into the KEGG REST API (kegg.jp/kegg/rest/keggapi.html) and couldn't find anything related to your question... $\endgroup$
    – mgalardini
    Commented Aug 21, 2017 at 8:17

1 Answer 1

1
$\begingroup$

Some information seems to be available on the module level utilising the API.

Reaction R00352 points to the module M00173. While some information is in the Definition, it's not trivial to link it to the particular reaction. This linkage is however easier accessible via API:

Reaction orthology defined on module level

$\endgroup$
2
  • $\begingroup$ Thanks. There are two issues with this as far as I can see: i) the third KO (K01648) is missing in the list and ii) it is not clear which logical relationship exists between these KOs; it seems it is always an AND but it is not entirely clear (e.g. from the link you provide: K00239,K00240,K00241,K00242,K00244,K00245,K00246,K00247 - do I need all of these or not?. $\endgroup$
    – Cleb
    Commented May 25, 2019 at 21:56
  • $\begingroup$ This table seems to point where the indicated members should all be required as one. In a couple of cases I looked it is indeed the case. Hence ad i) K01648 is separate so should not be noted ad ii) a very detailed decomposition to subunits, but yes. $\endgroup$
    – ellimilial
    Commented May 25, 2019 at 23:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.