Proposal: Bioinformatics

Bluntly put: why do we need a new site?

There are already

I imagine that every question that can reasonably be posted on Bininformatics would be ontopic on at least one of these sites, and that many bioinformatics experts are already active on at least one of these sites.

The latter is particularly important since you'd need these experts to shift their focus to a new site, which is a lot of inertia to fight against.

Note also similar discussions on the recent AI proposal.

share
2  
Note that there were proposals for bioinformatics sites before, but they were closed (and deleted by now). – Raphael Jun 12 '16 at 15:03

StackExchange is about building expert communities. No question asked on MathOveflow would be off-topic on Mathematics stackexchange but the sites still manage to draw questions that are very different from each other.

There's https://www.biostars.org/ which manages to be a thrieving community that discusses bioinformatics questions that aren't currently found on a stackexchange website.

I looked at the front page of Biostar and found Are there any software for the alignment-free bacterial whole genome sequence similarity search and comparison?. I think that question would likely receive no answer on any stackexchange website.

The hope of a bioinformatics stackexchange would be to have a website like biostars inside stackexchange.

share
2  
a) MathOverflow existed well before SE 2.0 and independently on SE. It also has a very specific research-level focus with its very own set of experts. So this is a bad example here; its key properties don't carry over to this proposal. b) If there already is biostars.org, what are the chances of establishing a new community or even move their experts here? c) You did not address my points. What makes you think you can move experts from sites they've been on for years to the new one? In particular, how to the existing sites not work for bioninformatics questions? – Raphael Jun 13 '16 at 12:39
    
@Raphael : The key question isn't whether you could ask a question on a website but whether the website succeeds in developing a community of experts. The biology stackexchange website and for that matter the cogsci stackexchange website didn't develop into very active experts community in a way that MathOverflow or Biostars did. I think that's largely to the amount of questions from lay people in both of those stackexchange websites. Their scope is too wide. – Christian Jun 13 '16 at 14:02
    
We don't see that on Computer Science; we've been battling scope erosing quite well. I don't see any indication that a scope narrower than one academic field of about the size of a department at your average university works well (on SE). Bioinformatics is, as far as I am aware, pursued by specialized subsets of all biologists and computer scientists, so you get (a lot) less than a department, and many of your experts may identify more as biologists or computer scientists, anyway. (You have yet to explain how you want to move people away from biostars.org.) – Raphael Jun 13 '16 at 14:07
    
But more to your point, I disagree. They key is whether a question develops a community of experts suitable for your questions. Which questions can not be answered by biology, compsci, scicomp, ... experts individually? And how many cross-discplinary experts are there in total? How many do you expect to gather on the site? – Raphael Jun 13 '16 at 14:13
1  
@Raphael : Most biology experts aren't on biology stackexchange. Real lay people don't ask questions about computer science on Computer Science because people the average person doesn't have real computer science questions. If a lawyer would post a question about a computer related concern to Computer Science stackexchange his question would likely be closed. That not true for the cogsci or biology stackexchanges. – Christian Jun 13 '16 at 14:18
    
1) Out of the top 16 users on Biology (this year), 7 state doing research in biology or similar, 5 seem to hold at least some degree in biology or a related subject, and 4 don't give any information. So while it's probably true that most experts are not on the site (which is probably true most if not all SE sites), the core usership seems to be quite expert. 2) The scope description of this proposal explicitly includes laypeople ("others interested in bioinformatics") so I'm not clear on how you want to make this a safe haven (?) for academics/researchers/experts. – Raphael Jun 13 '16 at 15:08
    
(Side note: we do get questions by laypeople on Computer Science, but it's true that they are few and far in between. Computer science has not yet permeated our culture in terms of common knowledge/understanding as a scholarley subject, even though its products are ubiquitous and central to recent technological advances and cultural changes. That's a big issue, imho, and I'd gladly entertain more laypeople on Computer Science if that meant CS had reached cultural assimilation.) – Raphael Jun 13 '16 at 15:10
1  
@Raphael : I grant that among the top users in Biology.stackexchange there are experts but the quantitiy is low. Lay people normally don't have bioinformatics questions just as they don't have Computer Science questions. But they do have cogsci, biology or politics questions. – Christian Jun 13 '16 at 15:17
    
Okay, now I don't get what your point is. There are biology experts on SE, there are CS experts on SE, there are statistics experts on SE. That means all the (types of) experts you need are already on SE. Is your real problem that you don't want to mingle with laypeople, or that you think you need and don't yet have bioinformatikcs experts specifically? You have yet to provide evidence that there are questions supporting the latter. – Raphael Jun 13 '16 at 15:20
    
Plus, so far I don't see bioinformatics experts flocking to this proposal. If you can gather enough, this proposal will move forward no matter what I say -- I just don't get the reasoning and expectations for this proposal. – Raphael Jun 13 '16 at 15:22
    
@Raphael : There are nearly no biology experts on SE. Even if you can point to 12 people that do research or hold a degree. The reason that the biology SE doesn't grow into something bigger is too much laypeople. – Christian Jun 13 '16 at 15:34
    
Yeeees, I get that that's your opinion. But how do you plan to make things different with this proposal? – Raphael Jun 13 '16 at 15:36
1  
@Raphael : Things are different than the biology stackexchange because the target group is narrower. Lay people generally don't think they have bioinformatics questions. – Christian Jun 13 '16 at 20:22
    
Let us continue this discussion in chat. – Raphael Jun 14 '16 at 8:17

Although I agree that most questions a bioinformatician might want to ask could fit in one of the stackexchange sites you mentioned, I sometimes have a feeling that some site more specifically dedicated to bioinformatics issues could be useful. I know that biostars exist, and I find it a pity that this site is not integrated in the stackexchange network. I got used to the stackexchange-like interface and still haven't gotten into the habit of following what is going on on biostars or ask my questions there. Maybe it's just me not having made enough efforts yet.

Now why the already existing stackexchange sites might not be enough to host bioinformaticians' questions and answers?

The kind of question I would see fit for a bioinformatics stackexchange site would relate for instance to the use of programs or libraries specifically used in the bioinformatics field.

Why not on Stackoverflow?

Such questions might not get the attention of the typical stackoverflow user, because they deal with specific tools not used in other fields. I'm not sure this is a valid argument, but it seems to me that questions easily get "swamped" in stackoverflow. Sometimes, an interesting programming question relevant to bioinformatics might get posted there, but go un-noticed because there are so many other questions being posted at the same time. That being said, the bioinformatics tag could help the bioinformacian potential asker or answerer.

Why not on Biology?

This community may not have a high enough concentration of people expert in the use of computational tools. It also may contain too many questions most bioinformaticians don't really care about.

Why not on Crossvalidated?

I don't have much experience with the other sites you mention, but I guess that the same phenomenon I suspect above for programming questions could exist for statistics questions. There are statistical tools or models specifically designed with biological data in mind. Questions relative to these specific tools or models might more easily attract interest on a site dedicated to bioinformatics.

These are my impressions. I'm not 100% certain yet that a bioinformatics stackexchange site is something absolutely necessary, but if it existed, I might use it.

Edit: I regularly encounter issues that I would like to ask on a bioinformatics stackexchange site. I gave some of them as examples here:

See also my comments to the present answer for examples of questions people asked on other stackexchange sites but would better fit, in my opinion, on a bioinformatics site.

share
1  
Here is an example of question that might fit better in a bioinformatics site than on stackoverflow: stackoverflow.com/questions/40475233/mauve-on-ubuntu-16 – bli Nov 21 '16 at 14:41
1  
Here is another one: stackoverflow.com/questions/40343740/… – bli Nov 21 '16 at 14:54
1  
Another one that could potentially fit stackoverflow.com/q/40747196/1878788 – bli Nov 23 '16 at 17:01
1  
Again: stackoverflow.com/questions/41465320/… – bli Jan 6 at 9:27
1  
And another one: stackoverflow.com/questions/41510048/… – bli Jan 10 at 13:54
1  
1  
    
Good point El Capitan – hello_there_andy Mar 11 at 8:31

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .