The Biosphere

arboviral

09:02

@canadianer Actually I think the most pressing problem then becomes the explicit overlap - the scope of any special-field site would fall completely within Biology.SE, which is grounds for closure of a proposal. This is what finished off the ecological statistics proposal. (I still don't really know how Bioinformatics.SE got through - any argument that you can make for some aspects of Bioinformatics not being within site scope could apply equally well to ecological statistics, really).

terdon

@arboviral Bioinformatics is explicitly off topic here.

Only the underlying biology can be asked here, while over there we also cover things like using software, installing, specific tasks that bioinformaticians need to deal with etc.

arboviral

@terdon The existence of 810 questions tagged bioinformatics would seem to contradict that. biology.stackexchange.com/questions/tagged/bioinformatics

terdon

This has always been off topic here.

arboviral

@terdon " any argument that you can make for some aspects of Bioinformatics not being within site scope could apply equally well to ecological statistics, really)" still applies too.

terdon

@arboviral I think you'll find all of those are about the biology.

@arboviral Oh, I don't think splitting up the site would be a good idea. But as a bioinformatician and active on both sites, I can tell you that there is actually little overlap between biology.se and bioinformatics.se

arboviral

09:06

@terdon In the first handful I get "https://biology.stackexchange.com/questions/65225/alternatives-to-ncbi-blast-du‌ring-us-government-shutdowns"

That really isn't about the biology.

Mad Scientist

the bioinformatics line was drawn somewhat arbitrary, it could have been included in the scope. Not sure if it would have been a good idea, mostly because specific communities seem to attract the right population more often than broad ones

arboviral

Well, okay, it's 6th. Polydactyly required.

terdon

@arboviral No, but I'd call that off topic according to my understanding of our scope here.

Mad Scientist

Biologists use bioinformatics tools very often, so that part is clearly on-topic here.

terdon

8

Q: Should bioinformatics related questions be considered off-topic?

I know there is biostar, which is specific to bioinformatics questions. However, I feel that there are quite a few parts of bioinformatics that have a lot of biological relevance. Perhaps we should draw the line at programming related questions?

> We should allow all questions that are asked from a biologists point of view, those are relevant to the users on this site and we should have the expertise to answer them. This would include using bioinformatics tools and understanding the principles behind those tools.

That's what I've felt was the case here.

arboviral

09:08

Biologists also do "things like using software, installing" that @Terdon mentions above.

terdon

@arboviral Yes, but that isn't on topic here.

In my experience, anyway

But [bioinbformatics.se] has a lot of pure programming questions, for instance, about using specific libraries or addressing specific bioinformatics problems.

Those wouldn't be on topic here.

arboviral

Most of what I do that isn't meetings or management, as a Biology PI, is programming these days.

terdon

Also, because by its nature the scope of the site is much more specific and opaque to no-experts, it became a site for experts from the beginning.

@arboviral I don't doubt that. I always considered myself a biologist too despite not having touched a pipette in almost 20 years now.

arboviral

Not disagreeing with any of this, obviously.

terdon

Personally, I would have loved to have opened this site to bioinformatics, but I always felt the consensus was against that.

arboviral

09:13

I still don't think we have a clear line here. And I'm not clear on what the "biology part of bioinformatics" is. If we're excluding methods, then why? We don't exclude mol biol methods, even though most are more accurately chemistry or physics. We don't exclude mathematical modelling. We supposedly cover ecology & evolutionary methods (I misremembered the proposal, but my question about it is here: biology.meta.stackexchange.com/questions/3421/…)

We're not being consistent. Either the EcolEvolMeth should never have been closed, or Bioinformatics should never have been approved.

@terdon I do some retraining every few years before they shoo me out of the lab. :)

terdon

@arboviral Well, my take is that if you're asking about the details of how, say, the blast algorithm works, that would be off topic. If you're asking about how to interpret your blast results and, in order to answer, you will need information about the algorithm, then it is on topic.

So if your question is aimed more towards the biological interpretation/application and less towards the nuts and bolts of the algorithm, it would be on topic.

Mad Scientist

the likely off-topic parts are e.g. programming questions regarding bioinformatics problems or using bioinformatics libraries. Simply using existing tools should be on-topic, modifying those tools or scripting them is a bit different

arboviral

But what about if you're asking whether a gamma distribution or a poisson distribution is the best way to describe the extrinsic incubation period of an arbovirus? That's equal parts stats and biology.

Not a great example, but just to illustrate.

terdon

I'd call that on topic. Probably. Once someone explained what it was about to me ;P

arboviral

Or whether ABC-SMC or ABC-MCMC is a better Bayesian fitter for insect population models?

Mad Scientist

09:16

one common way of making the line is to ask if you can remove all biology from the question, does it still make as much sense? Then it might be off-topic

terdon

Exactly

arboviral

Almost every question I can think of these days that I can't trivially answer myself is straddling 23 or more SEs.

Ah, now that might work!

Er, 2, not 23 :D

Mad Scientist

Personally, I'd prefer to have a Science.SE site for everything, but that would probably require a lot of things to be different, and probably wouldn't attract enough people

terdon

This is a good example of a question that's close to bioinformatics yet on topic:

8

Q: Finding proteins in DNA sequence

I have to do a task for a university task and I need to understand some things before figuring out how to do it. The task is the following: Find matches of known proteins (DNA-PolyI,II,III) to the specific E.Coli DNA, sequence. I downloaded in FASTA format the protein sequence of DNA-Poly3...

dna-sequencing

In my answer, I stopped short of going into the more "computery" bits and focused on the biology.

arboviral

We'd need better filtering, certainly. And that technically wouldn't cover stats (according to most statisticians I know)

terdon

09:18

> I will not explain all these options because that is beyond the scope of this site. Have a look at their documentation (which is quite good and clear) and if you still have problems, you could ask a question over at biostars.org.

Ah, time to change that link!

arboviral

Can we edit that into the site description/tour somewhere?

Maybe after a proper meta discussion/poll

Does that not exclude e.g. the occasional questions we get about debugging odd PCR results etc, though? (Mostly just playing devil's advocate; I think we could work it out...)

terdon

@arboviral I'd say that depends on whether understanding the problem requires understanding the data. For example, we often get bioinformatics stuff on Unix & Linux. If the question can be framed in such a way that any Unix expert can answer, then it's fine. If, however, it needs knowledge of a particular bioinformatics data format, for instance, I'd call it off topic.

Similarly, if the stats issue can be understood by pure statisticians and the biological nature of the underlying problem is not relevant, then I'd say it's best posted on another site.

If, however, you need to understand some of the biology to get the problem, then it belongs here.

@arboviral That's a good example, actually. You can't debug them without some understanding of the method and the chemistry involved.

canadianer

14:49

@arboviral My recommendation is closing bio and only having specialized sites.

arboviral

16:41

@terdon Wait, so you think thermocycler problems are a good example of something which should be on-topic, or off-topic? It sounds like the latter, but I'd expect that a (properly-researched) question on this sort of thing is pretty much the only sort of question a professional biologist would need to ask the community about via the SE format.

@canadianer The thing is, there is still a place for questions from the general public about biology, and there is still a duty on professional scientists to support that (literally, in the UK; if you are in receipt of BBSRC funding you're supposed to do a minimum of 2 days of public engagement a year, STEM ambassadors have CPD requirements, etc). I'd prefer Bio.SE to become more like a permanent AMA than close entirely.

If Bio.SE was pitched as a public engagement platform (like I said above, rebrand Bio.SE as BioCurious.SE or something) then you might attract a lot more users interested in developing public engagement skills. You'd still end up with a fundamental split between askers and answerers though, which isn't really how SE is supposed to work.

Unfortunately I don't really see a good solution, which is probably why I'm suggesting multiple bad ones... :D

@terdon Playing devil's advocate again here, but I can see situations arising where the biological aspect doesn't seem relevant to a non-expert, but actually is. The fact that they're blood samples is significant because the haem is inhibiting the reaction, that sort of thing.

canadianer

17:05

@arborviral I agree and welcome good questions from anyone.

terdon

@arboviral Hmm, I'd say on topic. That's a tool professional biologists use, so I don't see why not.

@arboviral Precisely. That's a case where understanding the underlying problem is relevant, so I'd want it here.

As for other questions 'experts' might have, I can imagine asking about experimental techniques, buffer preparation, extraction columns, all sorts of protocols and the like, for instance.

But also things that are simply not in your area of expertise. Say you work on social insects and now want to do some metagenomics on them. You know a hell of a lot about bees, say, but not much about genomics. Your question will still likely be of interest and be of a much higher caliber than that of a lay person.

That's kinda what I was hoping to see originally in bio.se. For example, I know a fair amount about genomics and gene evolution, but I've never done any work in epigenetics. Were I do move into that field, I can see myself asking on a site like this one wants to be.

2

Bryan Krause

17:23

Crossvalidated runs into a lot of similar issues with programming versus theory. Explicitly crossvalidated is supposed to be solely about statistics but there are definitely programming questions that sneak in - they are supposed to be removed but sometimes a well-asked question is approved on its other merits.

I kind of feel like bio.se is (or should be) similar: you have to have some content of biology. If you are really just asking how to do a programming technique, even if its a technique biologists use, its not on-topic

Bioinformatics successfully broke off from bio I think in part because they decided to explicitly allow more of those programming-based questions, that relate in particular to the tools used in bioinformatics

terdon

@BryanKrause Didn't really break off, as such. Off the top of my head, only two regulars there were contributors here.

No, 3. bli, Konrad and myself.

Bryan Krause

17:49

Well I meant more how that stack was able to differentiate itself from bio and not overlap with what's here

I think that would be more of a challenge with some of the other potential bio offshoots people are suggesting (biochem, mol bio, etc)

terdon

Yeah, I am not in favor of splitting off subtopics. All of those are on topic here.

And it's not like we have too many questions to handle, exactly.

Bryan Krause

19:06

Right

I think the notion that splitting into subtopics will increase participation in each is a bit dubious

The psych and neuroscience stack, for example, is a bit further from the scope of biology but at the moment is mostly just keeping some of the worst questions away from here, it's struggling quite a bit to get a foothold.

I just realized that in all the time I have lurked there I have yet to find a single question worth answering :-/

canadianer

20:16

To be clear, I don't really think splitting into subtopics would help participation either.

Mad Scientist

the site should be useful to professional scientists. If that doesn't work, it should at least be interesting. But the most successful Q&A sites are successful because they are useful.

2

Transcript for