Definition for interactions involving small molecules

[Chris Yates]

Hey all, 

I notice 6 unique interaction types involving gene and small molecule interactions. I am wondering if my way of defining these interactions is correct. If there is someone who is better informed than me, please let me know how you interpret these interactions. 

1. 'controls-production-of': 

Example: 

PARTICIPANT_AINTERACTION_TYPEPARTICIPANT_B

GeneXcontrols-production-ofCHEBI:15729

My interpretation is that GeneX uses small molecule reagents to catalyze reactions that produce CHEBI:15729. In other words, CHEBI:15729 is downstream of GeneX

2. 'controls-transport-of-chemical'

PARTICIPANT_AINTERACTION_TYPEPARTICIPANT_B

GeneY 'controls-transport-of-chemical'  CHEBI:46911

Simply that GeneY transports CHEBI:46911 from one area of a cell/tissue to another. 

3. 'chemical-affects'

PARTICIPANT_AINTERACTION_TYPEPARTICIPANT_B

CHEBI:46911 ' chemical-affects  '  GeneX

Many of these metabolites are exogenous, although not all. This makes me think that these are interactions that modify the thermodynamics or kinetics of an enzyme. 

4. 'consumption-controlled-by'

PARTICIPANT_AINTERACTION_TYPEPARTICIPANT_B

CHEBI:15729consumption-controlled-byGeneX

My interpretation is that CHEBI:15729 is used a reagent for a reaction that is catalyzed by the enzyme GeneX, and the utilization of  CHEBI:15729 results in its consumption. In other words, CHEBI:15729 is upstream of GeneX.

These two related examples are directly from the data, hence why I am confused. Is it correct to assume that GeneX does reversible reactions, and so GeneX is both upstream and downstream of the reactions? Or am I interpreting these incorrectly?

 5. 'used-to-produce'

PARTICIPANT_AINTERACTION_TYPEPARTICIPANT_B

CHEBI:16199 used-to-produce  CHEBI:32682

These are interactions directly between small molecules. It seems like these are declarations of what is a reagent for the production of what. This might exist purely for the sake of mapping reagent/product relationships or if there is relevant information for what gene catalyzes the reaction that is missing. 

6. 'reacts-with'

PARTICIPANT_AINTERACTION_TYPEPARTICIPANT_B

CHEBI:16199  reacts-with    CHEBI:32682

These seem to be direct interactions between metabolites without a description as to their nature. 

Please let me know if any of these are incorrect, especially my interpretation of 'controls-production-of' (which seems to be where the gene is upstream of the small molecule) and 'consumption-controlled-by' (where the small molecule is upstream of the gene). 

[Chris Yates]

Here is a more specific example that explains why I am confused by these two interaction types:

ARG1 converts Arginine (of which there are multiple related species here) into Ornithine and Urea. However, if my logic about the definition of the interaction types is correct, then Ornithine (with multiple species of it), Urea, and Arginine are listed as both reagents and products. I am also curious why there are so many of these extremely closely related small molecules that are treated as separate reactions. Although, I am more interested in understanding the correct definitions of the interaction types.

[Gary Bader]

Hi Chris - did you see the definitions for these at https://www.pathwaycommons.org/pc2/formats ?

Best,
Gary

Please let me know if any of these are incorrect, especially my interpretation of 'controls-production-of' (which seems to be where the gene is upstream of the small molecule) and 'consumption-controlled-by' (where the small molecule is upstream of the gene).  --
You received this message because you are subscribed to the Google Groups "pathway-commons-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pathway-commons-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pathway-commons-help/e9252468-5c93-4443-b600-6f151bb84e6an%40googlegroups.com.

[Ozgun Babur]

Hi Chris,

Your interpretation is mostly correct. Please also compare that with the descriptions at the link that Gary sent.

The reason for "Prot controls-production-of chem" and "chem consumption-controlled-by Prot" relations to exist simultaneously may be because of reversible reactions as you predicted. 

Please be careful with the terms "upstream" and "downstream" in your interpretations. You got them right in a technical sense, considering how this directed graph is defined. But if you are using these terms for the direction of biological signal flow, then there may be contradictions. For instance, "chem consumption-controlled-by Prot" means the Prot modifies the chemical. If we inhibit the Prot, the chemical will not be modified. So you can say the chemical is downstream of the Prot, even though it is upstream on the SIF graph. This means an analysis needs to consider the semantics of each of these relations rather than blindly following edge directions for the prediction of the effect of some perturbation. 

Best,

Ozgun

[Chris Yates]

I had not seen that, but that is absolutely perfect! Thank you very much!