/sci/ - Science & Math

>>14970524
>R
Best plotting package (ggplot2) and fantastic statistical packages out the ass. Rmarkdown knitr lets you write a dynamically generated paper, which is STEM's reproducibility wet dream (I work with some audit-trail necessary stuff, and I have it set so that every time you "open" the paper, it recalculates all statistical analysis and subsequent graphs from the raw data, including injecting the proper values directly within the text of the paper itself. Good stuff.)
Everything else is second or domain specific, imo

>>14970524
Python, Lah teks, any office suite.

>>14970524
wolfram mathematica, wolframscript, matlab

>>14970936
>matlab
Fuck off with the worst "programming" language ever made
>not free
>fucking 1+gb just to run
>Ridiculous number of datatypes, why the fuck
>Great packages for image analysis and others! Okay cool that's useful I'll just use th-
>Oh wait, they consolidated 10% of the functions I use into a newly named function, removed half of them, and renamed/changed inputs and outputs for the remaining functions I use
>This process repeats every version, which they release every 6 months
>Did I mention that the A and B versions released every 6 months have to be entirely redownloaded?
>Oh sorry did you want to work on more than one computer? Too bad lmao
Matlab is for academic grifters

>>14970951
just pirate it
works for me

>>14970896
There are two schools, one is R, the other is python/pandas/scipy/etc. R tends to be more in scienceland, pandas for techie data scientists. I'm actually a R/python guy, pandas features seem like a bastard child of python and R. Take your pick.

Consider C, not C++ in case you need to write something fast. Problem with C++ is you can learn C++ but not really learn it, it's a complex language with lots of ways of doing things. With C, if you learn it, likely code will be decently fast.

For word processor, whatever. I find most of the time people just use word, so word or libreoffice. I'm a big LaTeX/LyX fan, but where I am, in bioinformatics, word is typically used. It would be a waste of you learn LaTeX then your PI / coauthors / collaborators are all using word.

>>14970524
CAD
Statistical analysis
A big boy language like C, C++, .NET, or Java

Go forth and make whatever you wish.

>>14971048
Advice for getting into bioinformatics as a physics guy? Very familiar with pandas if that helps.

>>14972209
Gtfo, bio is full. You picked phys stay there or annoy doctors cause they are tyrants
>>14970896
When will they replace it? Fucking hate R and python cause i have to learn them instead of focusing on the triple C

>>14972275
Lol I’m watching genomics happen from the outside and I’m jealous that I might miss the biggest scientific revolution of our era. There must be room for just one physics refugee.

>>14972287
Not that guy but bio is only full of retards, we could use more physicists with a deep knowledge of probability, don't listen to that guy he's mean. I had a genomics professor at a top 5 public university who did his phd in physics, really smart guy.

Now I'll tell you how to get into bio if you tell me how to understand probability without spending 2 years learning real analysis and set theory.

>>14972287
You and mathguys are smarter than us and can easily replace us if u learn a few things, we also were waiting for this

>>14972331
>You and mathguys are smarter than us and can easily replace us if u learn a few things
SIR YOU DIDN'T NEED TO TELL HIM THAT IT'S A SECRET SIR PLS DO NOT REDEEM THE SECRET

>>14972330
It’s been a while since I read it, but I think Ross is quite good for probability without getting into measure theory, etc.

>>14972344
Might be a little too detailed but the first few chapters seem pretty fundamental, I'll try to work through them at some point, thanks. In terms of getting into biology, I would recommend starting with the literature; just check out recent issues of nature, nature communications, nature chemical biology, nature genomics, etc. and see if you can find some papers you enjoy (if you don't understand what they're about, the discussion section usually says what the applications in reality are). Find a topic or data analysis pipeline that interests you and you can find the details about them from the methods. Most modern bio stuff doesn't have textbooks about it because relevant methods change every year. There really isn't too much 'fundamental bio' knowledge in the same way as there is fundamental physics besides the central dogma and natural selection and all that.

>>14970936
Based, mathematica is the theorethical physicist's choice.

>>14970524
It probably depends on your use case and what the people you are working with are using. But not considering those factors my favorites would 1. Mathematica 2. Python and friends (sagemath) 3. Latex. These 3 should be able to cover every possible scenario a scientist needs.

>>14973219
What do you use sage for? I’ve been thinking of picking it up just for fun but I can’t imagine a use case outside of abstract algebra research.

>>14970524
Python for data/machine learning/plotting/statistics
C++ for high performance or hardware
Latex for writing papers
Hard to go more specific than that.

>>14974928
>Latex for writing papers
How do people use this? It's so unintuitive and seems like a time sink to learn when you can just use word

>>14975792
it's better than word in every single aspect but it has a steeper learning curve. once you learn it becomes just like writing in word.
it's standard in academia

>>14970524
[math]\TeX[/math]. Rest is task-specific.

>>14972209
Medical physics route

>>14975989
It depends of your field. If you have a lot of equations then latex is really worth it although word is finally starting to understand latex equation properly. In most field it is better to use word as you want a decent track change system with you coauthors.

My point is that latex is for formatting and the publisher takes care of it so I really don't see the reason to use it outside of writing equations.

>>14975792
It is highly autistic but allows for extreme control of the finished product

>>14970524
LaTeX, Coq, and Minecraft

>>14976081
Use Latex and Git to track changes.