Wikipedia:Articles for deletion/Protein-protein interaction prediction

The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result of the debate was no consensus to delete. W.marsh 21:13, 19 March 2006 (UTC)

[edit] Protein-protein interaction prediction

Original research doesn't belong on wikipedia, as per WP:OR Jude (talk,contribs,email) 03:31, 11 March 2006 (UTC)

• Delete per WP:NOR. --Rory096 03:32, 11 March 2006 (UTC)

*Delete as nominator. Jude <(talk,contribs,email) 03:33, 11 March 2006 (UTC)

• Massive Rewrite as per Thatcher131. Jude (talk,contribs,email) 09:12, 11 March 2006 (UTC)

• Keep-Update This is not original research it is the overview of a field of study. please explain how it(the article) is original if:
  • it has no "unpublished theories"
  • it has no "unpublished data"
  • it has no "unpublished concepts"
  • it has no "unpublished arguments"
  • it has no "unpublished ideas"
  • it has no "unpublished interpretations"
  • it has no "unpublished analysis'"
  • it has no "novel narrative or historical interpretation"

it is just a summary of what is published on the subject.help me make it more apparent if you wish Tim 03:40, 11 March 2006 (UTC)

• delete this looks like terrific original research. unfortunately, wikipedia is not the place for it. Niffweed17, Destroyer of Chickens 03:55, 11 March 2006 (UTC)

• Keep User:Tim@ Please give a specific argument for "original research" not just "I think it is". Quote something and I will find you a previous publication of your quote. 04:00, 11 March 2006 (UTC) —The preceding unsigned comment was added by Tim@ (talk • contribs).

• Keep"the only way to verifiably demonstrate that you are not doing original research is to cite reliable sources which provide information that is directly related to the article, and to adhere to what those sources say." The article does this, therefore this page does not violate WP:OR and should stay. "Wikipedia is not an experiment in democracy. Its primary method of finding consensus is discussion, not voting. " upon inspection of http://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion/Precedents#Original_research you will find that the page in question is not Original research. And it does belong in Wikipedia. Until someone says something intelligent or any argument at all in opposition I will try to keep the page --Tim 04:44, 11 March 2006 (UTC)

I'm sorry if you are upset or offended but if you are unwilling to recognize the opinions of others and work with them you will find a hard time being accepted here. And, frankly, if the consensus goes against you, the article will be deleted whether you like it or not. Your best bet is to participate with an open mind and at least a little respect for more experienced editors.Thatcher131 05:26, 11 March 2006 (UTC)

• Delete WP:OR and WP:NOT. Royboycrashfan 04:59, 11 March 2006 (UTC)

• Delete as original research. dbtfz^talk 05:02, 11 March 2006 (UTC)

• Keep I fail to see how this is original research. It cites a great deal of secondary sources and it is not making any unfounded claims or analysis. I do not know much about the subject other than it exists and this is a valid area of research. kotepho 05:08, 11 March 2006 (UTC)

• Massively rewrite and cleanup. Tim@, to be honest this looks like a college project or a first attempt at writing a review article for a journal. It's not original research but it is highly inappropriate for Wikipedia. Wikipedia is not Bioinformatics. Articles here should be intended for a general audience. Ideally you would explain in terms a high schooler or college freshman could understand, 1) what are protein-protein interactions, 2) why it is useful to be able to predict them with a computer, and 3) the general principles involved. The only wikipedia article I have ever seen with 150 references was on Rathergate, and that because it is so controversial. You should aim for no more than a dozen, preferably review articles from easy to read journals like Current opinions in ______. You also need to review the Manual of Style and become more familiar with writing wiki-like articles. My recommendation is that you voluntarily withdraw the article and copy it into your user space (see WP:USER.) User pages are a great way to work on a complex article or figure out how to write a wiki article and they are not subject to (most of) the content guidelines of main article space. Thatcher131 05:23, 11 March 2006 (UTC)

• rewrite Thatcher131 thank you for the feed back. I will edit the page so that it is more to your liking. I thought it was no less for advanced than other Wikipedia pages I have found of use: http://en.wikipedia.org/wiki/Support_vector_machine for example. I will work on making it more palatable for a "general audience". You are welcome to help.--Tim 05:41, 11 March 2006 (UTC)

I certainly don't understand Support_vector_machine either but it follows the general principles better, opens with a general summary; gives some specifics, and ends with a representation of the literature.Thatcher131 07:04, 11 March 2006 (UTC)

• Delete per Thatcher131, who didn't vote delete, but I agree with the principles, if that makes sense. Tim, it looks like you have a broad perspective on the literature for a field that isn't covered by Wikipedia well enough. Rather than try to add all your knowledge at once, why don't you start with smaller articles on individual methods? The narrower the topic of an article, the easier it is to avoid original research while tying it together. Once you've built such a foundation, it would be relatively easy to write an article named "Protein-protein interaction prediction" that links to the more specialized methods and draws on only a couple of secondary reviews. Melchoir 06:19, 11 March 2006 (UTC)

• Delete per Niffweed17. KillerChihuahua^?!? 06:31, 11 March 2006 (UTC)

• Delete as original research, and recreate in future. --Terence Ong 07:23, 11 March 2006 (UTC)

• Keep Not OR, was briefly mentioned in one of my biochem lectures. Just because it says prediction, dosen't mean the article is a prediction, but rather the process by which they form said prediction. Mike _{(T C)} 07:47, 11 March 2006 (UTC)
  • It's possible to write OR on any topic, is it not? Melchoir 08:15, 11 March 2006 (UTC)

• Keep but rewrite and cleanup per Thatcher131's most excellent advise. (I don't agree at all with Melchoir's suggestion of starting with smaller articles on individual computational methods.) Lambiam 09:04, 11 March 2006 (UTC)

• Rewrite as per Thatcher131 Dlyons493 Talk 11:16, 11 March 2006 (UTC)

• Keep per Thatcher131 Eivind 11:21, 11 March 2006 (UTC)

• Comment. This very clearly is not original research, but it may be over-complex for WP. There are many sicence articles that are equally complex, yet they do not have the number of references that this article has. It looks as if the author is prepared to learn how to make it acceptable, so why not withdraw the nomination for now and let him do it. I am only on the fringe of this sort of computational science, being more a small-molecule quantun computational chemist, but I do know this is important and it needs an article. Do not bite the newbie. Let him have a go. So Keep. --Bduke 11:39, 11 March 2006 (UTC)

• Keep. It requires a massive rewrite, but it meets Wikipedia:Notability. —Eternal Equinox | talk 14:01, 11 March 2006 (UTC)

Update help make this a good Wikipedia page by commenting at User:Tim@/PPIP--Tim 15:28, 11 March 2006 (UTC)

• Weak delete I think it's beyond the horrible point of no return, but I'm a cynic. -- getcrunkjuice^contribs 17:37, 11 March 2006 (UTC)

Question? is this issue settled, can i replace the delete box with a {{cleanup}}--Tim 19:56, 11 March 2006 (UTC)

• Keep This is definitely not original research. Perhaps a little bit of cleanup/organization needed, but completely appropriate. OhNoitsJamie^Talk 22:54, 11 March 2006 (UTC)

• Keep, it needs a massive cleanup for a general audience. Not OR. Tim, with a bit of editing, it might make a good review article for peer review publication -- Samir ∙ (the scope) 23:55, 11 March 2006 (UTC)

I will replace the delete box with a {{cleanup}} in one day unless someone objects?--Tim 03:33, 12 March 2006 (UTC)

I'm objecting. You can not do this. It is against WP policies. The consensus is probably for keep, but you have to wait for an Admin to close off the debate, unless you can convince the nominator to withdraw it. --Bduke 03:52, 12 March 2006 (UTC)

Ok i did not know that is how it works; thanks for filling me in--Tim 13:44, 12 March 2006 (UTC)

• Delete references are arbitrary, does not acknowledge much better already existing wikipedia pages on the same subject and does not add any content, also a alot of unsubstantiated and frankly wrong points of view e.g. folding docking methods unworkable, clearly the work of a tendentious amateur Zargulon 07:41, 13 March 2006 (UTC)

• Keep, but massively rewrite and cleanup. Great advice from Thatcher131. It would be inappropriate to delete this since it does not seem to be original research but writing on a topic at too high a level; that is an issue for clean up, not deletion. JGF Wilks 11:56, 13 March 2006 (UTC)

• Delete unless cleaned up or rewritten. Currently appears to resemble original research. Stifle 23:50, 13 March 2006 (UTC)

note: Zargulon the article states that folding docking methods are currently unworkable on a genome wide scale because the best algorithms from CASP are given a month to find one fold. Taking 50,000 months to fold the proteins of a genome and then 2500000000 months for docking (assuming the same time frame for a single prediction) after only 7 days to sequence it is to slow for use.--Tim 02:54, 14 March 2006 (UTC)

It is not because the CASP experiments give a month that the servers take a month to make a prediction, with the threading techniques a model is built in half an hour to a few hours, this varies with the length of the target sequence and whether homologues are known or not. If there is no known homologues, only ab initio techniques can be used and are restricted to small sequences, although the fragment approach is being developed with promising results. In addition, you are assuming that the genome wide experiment would be done on one machine, whereas this would be a massively parallel work. The servers involved in CASP are often clusters that are able to work on several targets in the same time. The limitations are more from the theoretical point of view, e.g. better potential energy functions, better detection of remote homologues... By the way, I can't find where the article states that folding docking methods are unworkable because of the time used for folding prediction algorithms, there is no mention of CASP either? Blastwizard 05:48, 19 March 2006 (UTC)

I am staring at that section of the article, and it does not state what you just said it states. Rather, its implication is that folding and docking will always be unworkable. In fact is so badly written that I cannot begin to suppose what the intended meaning may have been. It also incorrectly categorizes docking and folding as "dynamics methods" - short for molecular dynamics? or dynamic programming? either way it's wrong. Most docking and folding methods do not use either technique. Even if you felt that truly representative literature search was beneath you, how much trouble would it have been to inform yourself by taking a quick look at the Wikipedia articles on protein structure prediction and protein-protein docking? Zargulon 04:19, 14 March 2006 (UTC)

I hope the recent changes make the article more clear to you. You can of course contribute a solution as well as pointing out a problem. in any case, thanks for the help. --Tim 17:16, 14 March 2006 (UTC)

They don't. All you did to the docking/folding section was wikify it.. how is that supposed to make it "more clear"? I have already contributed a solution. Deleting the article is an excellent solution, because it will disabuse people of the wrong belief that there is a meaningful entry on protein interaction prediction, so that maybe some people will write one. It is not constructive to take an important page and fill it with gibberish. If you can't do any better, you should wait patiently for someone else to. Zargulon 17:55, 14 March 2006 (UTC)

Well I agree to disagree with you on that point because I believe nothing is ever perfect and everything is a work in progress and should be respected as such until updated with a better version. I believe this article provides at least some hints as to what PPIP is, how it works and how to find more in-depth information on the topic. I can not see that a blank page (however perfectly blank) is better than what information is provided. --Tim 07:37, 16 March 2006 (UTC)

No, there are a lot of things which do not deserve respect. Respect has to be earned. Please allow this page to be deleted and started from scratch.. that will earn you my respect. Zargulon 10:48, 16 March 2006 (UTC)

• Delete. This article is full of errors, inaccuracies and dreams from the author. Protein-protein docking does not work with predicted models but with structures solved by X-ray crystallography, if the rest of the article is the same type of fantasy, there is a lot of cleaning up. The description of threading for protein folding prediction is a bit funny. The sequence method described assumes the contiguity of residues involved in interactions, if patterns were to be found. I also object the wrong use of many terms, for example the active site is certainly not the site of interaction between two proteins. The best we can do is predict the residues that may be involved in interactions, providing that both the structures are known the rest is pure speculations. So far I don't know anyone who knows about proteins who would venture yet in such a hard problem. What is the point of an article on a method that is unworkable? Wikipedia is not for original research, there are peer reviewed journals for that Blastwizard 01:03, 17 March 2006 (UTC)

{{sofixit}} There are plenty of secondary souces to be found to support anything that is not original research. Cleanup is not a reason for deletion and neither is it being unworkable. kotepho 09:15, 17 March 2006 (UTC)

That is disingenuous. This article doesn't have to be cleaned up, it has to be started again from scratch. This is best indicated to potential editors by deleting it. Zargulon 10:06, 17 March 2006 (UTC)

I don't think fixing this article is a good idea, sometimes when a car is not roadworthy, the best thing is to scrap it and get a new one; that's my medication for this article. May I suggest instead that someone adds a few lines on the perspectives offered in protein-protein docking (but no more than a few lines) that is all there is to say. Blastwizard 14:12, 17 March 2006 (UTC)

• Delete; While this is a valid topic, the concerns raised by Blastwizard are troubling, and worse, Tim@ (talk • contribs) has resorted to minor-league talk page spamming to bring his article to peoples' attention. In the best manner possible, I'm trying to avoid WP:BITE because it seems he's making a genuine attempt to contribute in an area of his interest, but with the more speculative nature of this article than some may realize it falls under OR.  RasputinAXP  c 01:05, 17 March 2006 (UTC)

• Keep; Important topic. rewrite and cleanup will happen as a process of the articles evolution (As other users see this article, and edit it). --hydkat 06:23, 17 March 2006 (UTC)

• Delete unless it undergoes major revisions per Blastwizard. The use of references is rather odd. A lot of the papers seem to be completely unrelated, remove every thing that doesn't relate directly to the text (e.g. the first 3 papers listed). Tim, I think the first thing you want to look at is to make specific citations in the text, to references you think support your claims. I might want to have a look at some of the things I think look sketchy, but I don't want to read 30 papers just to edit this page. I think you have to be careful not to confuse docking of experimentally determined structures with docking of proteins without a known structure. I would be highly critical of results from the latter, so this requires good references! Kjaergaard 06:44, 17 March 2006 (UTC)
  • Comment please see protein-protein docking and make sure this is not what you are thinking of. Kjaergaard 07:01, 17 March 2006 (UTC)

• Keep per Bduke - he works in the field and knows the area, and says the article makes a valuable contribution; it is VERY well referenced, and the allegation that it is original research does not seem to be substantiated (from the perspective of a literate non-expert). It does need to be cleaned up a great deal though. --SilverWings 22:59, 17 March 2006 (UTC)
  • What makes you think he works in the field? This article should be deleted not because it is original research, but because it runs the gamut between highly misleading and completely incomprehensible. Zargulon 23:36, 17 March 2006 (UTC)
  • The author has no clue about the field nor has he a clue about biology. If it was an exam paper it would have a very bad mark, if it was a scientific publication it would be rejected by the reviewers. I have looked at the article in details and made my comments are in the article talk page. In addition, looking at the history of the article, it looks like it started as a copy from a book, was somewhat well written and well referenced but obviously written by someone else. Then it was edited to this incomprehensible gibberish. Blastwizard 14:35, 18 March 2006 (UTC)
• Keep. Does not seem to be primarily original research at all, many sources cited. Tag for cleanup. Herostratus 03:45, 19 March 2006 (UTC)

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.