/sci/ - Science & Math

Hm, just hearing of this now. Bookmarked, thanks op

>>6243800
this

this seems interesting

This actually looks pretty cool. Thanks man.

This is so shit, its supposed to be a bioinformatics site, but the DNA helix they use for the level meter is completely inaccurate.

>>6243858
>hey guys, look at how autistic i am! am i cool yet?

>>6243875
>autistic
>cool
wat

>>6243796
why the fuck would they use python for this?

>>6245658
What do you mean? To solve problems, or to program the website?

Finally! Finished all my population dynamics problems.

>>6245658
Why wouldn't they? Python is a widely used language in bioinformatics. It may not be as fast as C, but it has a better learning curve and development is quicker. It's more appropriate for scientists because you can get into it very quickly.

>>6245854
Yup, Python is probably the goto language for bio in absence of specific preferences. It has an enormous amount of libraries and material available, it's not proprietary, and the majority of other scientists use it.

It's also pretty fast with the right set up. Speed isn't really a huge deal in biology - it's not like in vidya where a 20% decrease in performance can be the difference between playable and unplayable framerate. If your program takes 10 times longer in Python, you just leave it running overnight, and enjoy the weeks of work you saved by skipping C (which would be downright painful to use because of how little sugar it has).

Also Python seems to be less confusing for non-programmers, so its a good choice for teaching.

>>6245872
Other popular languages seem to be:

- R
- Java (established programs, like heavy duty webserver software, seem to eventually gravitate towards being rewritten into Java by professional software developers)
- Matlab (awful string capabilities make it very unsuitable for sequence related problems)
- Perl (now falling out of favor because of python)
- Mathematica (don't like this one myself, but I see a lot of cute one-liners with it, if you're into that sort of stuff)
- Ruby (not sure why)

I see a lot of Lisp and Haskell solutions on Rosalind - although that's probably Haskellfags trying to find a use for their new language. I rarely see Haskell actually used in Biology.

>>6245875
Didn't read the rest of the thread, so sorry if the answer is somewhere in there.

I'm a 2nd year chemistry student, and I really want to learn some basic modelling skills. (Think basic energy transfer systems and heat distrution/movement). It doesn't have to have any visuals, only numerical output.

I'm fucking around with Mathematica, but is that to basic? I like the fact that I can learn Mathematica on the fly, with the great help function. Are there any other programming languages you can learn/work with on the fly that are better/more powerfull at basic modelling/simulations?

My university doesn't have any programming courses for chemists/bio on a bachelor level I can take.

>>6245880
Oops, didn't see that the thread is short as fuck

>>6245875
Pretty much true. I don't see that many people sticking to Ruby these days, though, mostly because it's not very good. Only real use I can see for it is if you're making web applications, since it has a pretty good framework.

Lisp and Haskell seem to me like those niche languages that are good but mostly only if the rest of your lab also uses them (which is rare).

>>6245880
Matlab is probably the best way to go about modeling, but the price may be too steep for you. Mathematica is good, stick to it if it works for you.

>>6245880
>basic energy transfer systems and heat distrution/movement
Look at the engineering faculty, especially mech and chem. They do that a lot.

>doesn't have to have any visuals
Matlab and Mathematica are both great for visualization though.

>Mathematica, but is that to basic
No, actually it's probably the most sophisticated tool on the market. If you ever feel like you're rowing upstream, try Matlab (or Octave). Beyond that, your options are python (which is like matlab but a bit more general purpose), C# (which is really strict, has many great intellisense tools, reasonably fast) and C or C++ (painful to use but you could be that one sought after guy who knows how to do super optimized, super fast simulations, so it's a great skill for your career).

>I can learn Mathematica on the fly, with the great help function
Yes, that's very nice. Never waste time on poorly documented languages.

>My university doesn't have any programming courses for chemists/bio on a bachelor level I can take.
Take general course. Ask the prof about taking the graduate courses (they're probably 101 level courses with 500-code anyway). Use online tutorials/courses/etc. Try to implement simple algorithms from papers or computational chemistry textbooks.

>>6245891
From what I've seen of Haskell, it seems really nice. Some features are really nifty, and its a very sophisticated functional language (unlike, say, Lisp, which is a joke - it's like handing someone a very sturdy stone axe and telling them to build a space ship).

The only reason I haven't learned it is that it seems really hard (I think it just has a front-heavy learning curve because of all the new concepts you don't grow up using, unlike procedural). The community is also not so big, but on the other hand most Haskell coders are very good programmers with very good understanding of computer science, unlike say Java or C# where it's Indian coders galore.

I think Haskell wouldn't be so bad for science. Most research software is written by one or two people and rarely extended per se (if it becomes a hit the algorithm is reimplemented by real coders, and by then you've moved on to better projects). Haskell's paradigm lends itself nicely to how people often think about mathematical and bioinformatics problems.

>Matlab price
BTW, any school that does any modelling at all will have an institutional Matlab license. You can try to track down the department that has it and ask to use it.

>>6245898
You do have a point about Haskell. I don't have much experience with it myself, really.

Of course, pricing is not a problem if you're connected. But as a student it's not always that easy (although they do give generous discounts for students).

>>6245900
Yeah, I have only tried the first few chapters LYAHG before giving up. So far, probably my favorite thing is how there is no drawback to making recursive functions - since every other problem seems to be solved through a process of "come up with a recursive solution, then figure out how to make it not slow as molasses" this is basically cutting half your work, and what's better, it's cutting the boring half.

This gets pretty challenging pretty quickly. Useful, since my programming experience isn't that great. Good recommendation, OP!

>REAR
god damn it motherfuckers put an NP hard problem right at the begnning, what the fuck

>>6243796

My level? ZERO! And proud of it.

Why waste your time (which we don't have much) on such a shitty field?

>A farewell to bioinformatics
>Published: March 26, 2012
>Category: biology:nontechnical

>I’m leaving bioinformatics to go work at a software company with more technically ept people and for a lot more money. This seems like an opportune time to set forth my accumulated wisdom and thoughts on bioinformatics.

>My attitude towards the subject after all my work in it can probably be best summarized thus: “Fuck you, bioinformatics. Eat shit and die.”

>Bioinformatics is an attempt to make molecular biology relevant to reality. All the molecular biologists, devoid of skills beyond those of a laboratory technician, cried out for the mathematicians and programmers to magically extract science from their mountain of shitty results.

Whole thing: https://web.archive.org/web/20120604093647/http://madhadron.com/?p=263

Kids, don't waste your life on bioinformatics!

>>6246854
10/10 A++++++ MAD would read again thank you

>>6246854
>bioinformatics is too hard for me

>>6246874
bioinformatics is mostly string matching and that shit ain't hard, you dumbass.

>>6246878
you're a pleb

>>6246878
And yet it was still too hard for that guy, who, instead of trying to improve the field, just ran away leaving a petulant note on his own blog.

>>6246887
I don't think he was in it to run away in the first place. He says he only has a Bachelor's.

>>6246886
>you're a pleb

uh, oh… that sure told me.


>>6246887
>And yet it was still too hard for that guy, who, instead of trying to improve the field, just ran away leaving a petulant note on his own blog.

too hard? it was a fucking waste of time and waste of his life.

he was slaving away for no benefit at all.

I just finished my frst semester of molecular biology, should I quit? It's not that interesting.

>>6246895
>It's not that interesting.
Yes.

>>6246895
>"It says here you originally started out as a molecular biologist, but decided to quit and change fields. Why was that?"
>"Oh, I asked on this site called 4chan if I should quit, and they said okay because biology is a pseudoscience anyway, so I quit."

>>6246902
>>6246907
Thanks guys, 4chan's 2 cents was enough to make me drop out. Now on to be a bum.

>>6246907
>they said okay because biology is a pseudoscience anyway
I've learned from /sci/ that any field with real world applications is a soft science.

>>6246895
go int o a field that has job prospects. molecular bio has next to none. and since you find it boring, you won't even be able to do it for pure love.

>>6246917
Then there's Geophysics... on the forefront of modeling weather patterns with accuracy, predicting earthquakes, and volcanic erupitions. Applying maths in ways you couldn't imagine with your puny mind

>>6246931
>modeling weather patterns with accuracy, predicting earthquakes, and volcanic erupitions
A-are you being sarcastic?