Talk:Rhodopsin-like receptors

From Wikipedia, the free encyclopedia

	MCB Portal This article is within the scope of the Molecular and Cellular Biology WikiProject. To participate, visit the WikiProject for more information. The WikiProject's current monthly collaboration is focused on improving Restriction enzyme.
Start	This article has been rated as start-Class on the assessment scale.
Mid	This article is on a subject of mid-importance within molecular and cellular biology.
Article Grading: The article has been rated for quality and/or importance but has no comments yet. If appropriate, please review the article and then leave comments here to identify the strengths and weaknesses of the article and what work it will need.

About deletion of octopamine receptors... Remember that Rhodopsin-like GPCR include a lot of invertebrate receptors, including invertebrate opsins and squid retinochrome (that is even not a receptor).Biophys (talk) 01:01, 10 December 2007 (UTC)

As a practical matter, I decided to limit the list to human class A GPCR so that the recently generated PBB GPCR pages could be rapidly integrated and to keep this page from becoming too big. Once the over all organization of this page settles down (see below) we can consider expanding it to incorporate receptors from other species. But even then, it might still be better to have two separate pages. One restricted to human receptors for brevity and a second including receptors from other species. Boghog2 (talk) 07:26, 10 December 2007 (UTC)

Yes, I am talking about family/class A! It includes more non-human proteins than human ones. There is clear and certain consensus in the scientific literature what is called "family A", and we can not reconsider this, as it would be contrary to WP:NOR policy. Saying that, I am not pushing anything here. This is just my friendly suggestion/reminder.Biophys (talk) 18:01, 10 December 2007 (UTC)

To extend your logic, any of the thousands of individual protein pages, including the vast majority of GPCR pages, that only discuss the human protein (or in the case of PBB generated pages the human + mouse orthologs) is contrary to WP:NOR policy. Hmmm. This WP article is already getting very long. In addition, 100% of the links on this page are to human proteins. So I really don't understand your objection to my limiting the coverage at least initially to human proteins. We can add proteins from other organisms to this or another article sometime in the future. Boghog2 (talk) 23:33, 10 December 2007 (UTC)

Sorry, I only suggested to keep the existing InterPro link to octopamine receptors.Biophys (talk) 00:18, 11 December 2007 (UTC)

Likewise, sorry. I have reinserted the octopamine receptor. One of the reasons I initially deleted it is that it wasn't clear to me where to insert it. After digging a little further, it is now clear to me that the octopamine receptor (and ligand) is very closely related to dopamine. Boghog2 (talk) 19:42, 11 December 2007 (UTC)

[edit] Wrong classification

Thank you for your great work, Boghog2! But unfortunately you picked up a poor source. I do not know who these Joost and Methner are, but their classification is wrong. Look at family 16 for example: it combines adenosine and histamine receptors. This is simply ridiculous: histamine H1 and H2 receptors belong to different 15th and 16th families! No wonder, a lot of InterPro families remained unclassiffied by these authors. What kind of classification is that? Also, no wonder that authors did not even bother to compare results of their work with other well-known classification systems. Please take a look at GPCRDB: [1]. That is more or less correct and widely accepted classification , which is based on analysis of alignments of many thousands (not 244) amino acid sequences.Biophys (talk) 01:46, 10 December 2007 (UTC)

First of all, thank you for staring this page Biophys! I think it is important to have page devoted to the organization of this large family of proteins. Concerning the classification, I stand by the one published by Joost and Methner (J&M). It is important to keep in mind that there is often more than one "right" way to classify proteins (e.g., function, sequence, structure, or some combination of these three) and the best classification depends on the intended application. The grouping and nomenclature of the histamine H1-H5 receptors is based on pharmacology (i.e., their ability to bind histamine). But that does not necessarily mean that these receptors are most closely related by sequence. I have added a published citation to the GPCRDB (Vriend et al.) to this article. I don't know if this is the best citation describing how the GPCRDB classification system was derived, but it is the best one I could find. From the Vriend publication, it appears that the GPCRDB classification is based soley on sequence alignments. However on closer inspection it appears that the GPCRDB classification may not be based entirely on sequence (for example, if the classication were based soley on sequence, why would the authors post this warning). So the GPCRDB classification may be biased to keep named receptor families together (and please note that this is not necessarily wrong, just different). The methods used by J&M are carefully documented in their publication and that is why I choose to base the organization of this WP article based on their work. Also it is important to point out the most complete/exhaustive analysis is not necessarily the best. In the context of this article where the 100% of the current internal WP links are to human proteins, the J&M classification is arguably the most appropriate. Boghog2 (talk) 07:26, 10 December 2007 (UTC)

Yes, Vriend's classification is mostly (but not exclusively) sequence-based, and yes, a "natural" biological classification (that is classification reflecting true divergent evolution of species) can never be only sequence-based.Biophys (talk) 18:53, 10 December 2007 (UTC) (of course here we are talking mostly about protein Paralogy, so this is divergence of genes/proteins rather than divergence of species).Biophys (talk) 19:01, 10 December 2007 (UTC)

RE: a lot of InterPro families remained unclassified by these authors. This is merely a consequence of the scope of the analysis. J&M chose to limit their analysis to human class A GPCRs and therefore by definition, it will not be possible to map some InterPro families on to the J&M classification. Just to clarify, J&M did not link their classification to InterPro families. These were linked by myself manually restricting myself to the links provided by you in the original version. I should be able to add more later when I have time. Boghog2 (talk) 07:26, 10 December 2007 (UTC)

Of course, the ligand-based or "pharmacology-based" classification in general does not necessarily coincide with sequence similarity-based classification. But this approach works well for this particular protein family, as far as I know. Authors have limited their sequence analysis to paralogous proteins from only one organism to bulld a sequence-based clasification. But this is a terrible error, especially within a family with barely detectable sequence similarity, such as many members of A GPCR family! Why? Because using sequences from multiple species allows a profile-aganist-profile type sequence comparisons (as Vriend did) rather than comparisons of individual sequences. By reducing the number of sequences, authors dramatically reduced the reliability of all their conclusions. But I do not know how well you are familiar with biological systematics and sequence analysis...Biophys (talk) 18:29, 10 December 2007 (UTC)

I wonder what kind of mess would be Pfam and InterPro if their authors did only comparisons of human proteins. This only means that Joost and Methner had no idea what they are doing, unlike Alex Bateman, Vriend and others who did a lot of good work in this area.Biophys (talk) 18:47, 10 December 2007 (UTC) Biophys (talk) 19:01, 10 December 2007 (UTC)

I think you are really blowing things out of proportion. If you take a very close look at the amine GPCRDB phylogenetic tree, it appears that the adrenergic and dopamine receptor subtypes at least partially overlap, but the differences in distances are quite miniscule. Taking this tree literally would imply the GPCRDB classification that treats adrenergic and dopamine receptors as distinct subgroups is in error. This is additional evidence that the GPCRDB classification has been slightly tweaked in order to keep receptor subfamilies together. Please note that the very close relationship between adrenergic and dopamine receptors is recapitulated in the J&M classification where both receptor groups are placed in subfamily A15. You apoint out that the muscarinic receptors M1 and M2 are placed in subfamilies A16 and A15 respectively and therefore conclude that the the J&M classification must be seriously flawed, even though the distance between these two subfamilies (as shown in Figure 2 of the Joost and Methner publication) is relatively small. The bottom line is that both classifications systems are in close agreement. Differences where they exist are relatively minor. Boghog2 (talk) 23:33, 10 December 2007 (UTC)

I am not suggesting to interpret individual trees. That would be over the top. I only suggest to use this overall GPCR classification. This overall classification treats adrenergic and dopamine receptor subtypes as two distinct subfamilies of "amine receptors", which is correct as far as I know, although if you are a good GPCR specialist, you probably know better.Biophys (talk) 00:13, 11 December 2007 (UTC)

No problem. The two classification schemes are actually very close to each other. Furthermore the sequence difference between the adrenergic and dopamine receptor subtypes is very small and the question as to whether these are included same or closely related subfamilies becomes a matter of semantics. I am now trying to fill in the missing pieces of the J&M classification with information derived from GPCRDB classification. I have also moved the histamine H2 to the end of subfamily A15 and the histamine H1, H3, and H4 to the beginning to subfamily A16 to emphasize the very close relationship between these receptors. Boghog2 (talk) 19:42, 11 December 2007 (UTC)

This is great. Just do whatever you think is best.Biophys (talk) 20:35, 11 December 2007 (UTC)