Java or Python for Bioinformatics?

Neither.

Java or Rust. Dear god not Python

Java

@12bitfloat any link to suggest me to look at, so far I came across BioJava but I am still quite not understand.

Matlab

I SAID IT OK, I FUCKING SAID IT

@wonwon0 yeah, shut up.
@johnmelodyme yeah dont do python. It will give you bad bioinofrmatics.

Science likes py
business likes java

but that's just an observation. Not a suggestion

Use whatever suits your application and has the better libraries. You'll get a lot of opinions on syntax and stuff here, those are important sure but ultimately irrelevant compared to package ecosystem. The idea here is to get *bioinformatics* work done as efficiently as possible, if that involves working with a language whose syntax you hate I think it's still worth doing.

Having clean syntax or program flow and missing your deadlines because you couldn't get your application working on time is pretty pointless.

Personally I would go with Python considering how popular it is in scientific circles. I've seen a lot of academics use Java too, so that should be a viable option if you can find a good library. MATLAB's inbuilt toolboxes are great and are massive timesavers, use them if you can get access to them somehow.

Depends on specification needs.

Long term, slower, and better performance - Java

Quick and dirty - Python

Scratch. Always scratch. 👍

I worked in a phylogenetics lab, a pharma lab and a material sciences lab.

The code I encountered was overwhelmingly Python and Perl, a bit of Java, and some R.

Python is a great & flexible language, it shines not because it's the best tool for the job, but because it's a tool that so many people can use.

At the latter two companies, they were actively transitioning to Haskell — which was the reason I got hired, but I wouldn't necessarily advise it. The language is amazing for the context, but the package ecosystem and hiring pool make things difficult.

It's all about balance... If hiring a team and prototyping new code is important, pick Python. If robust, trustworthy efficient code is important, pick Rust.

Most importantly though, use a language you're comfortable with.

It depends:
Quick and dirty data formating? Python
A lot of data to process but you dobt have to change the code?
Java, C#, (C and C++ if you are crazy enough)
You have too much money on the hardware and are ready to get crippling depression?
Mathlab
Other shit like: Rust, Kotlin, R if you want to be a fancy hipster because you are not mainstream.

Just use a lang you are the most comfortable with, if that is too slow, learn a better one or throw more hardware at it.

Why not create a bioinformatics-centric Javascript framework?

(you may downvote me to oblivion, I probably deserve it)

@wannabe yes you do

@wonwon0 are you trying to get murdered?

@Gregozor2121 R is actually a lot more mainstream in science-related fields than C#.

@bittersweet R is specifically for statistics so any field which requires data analysis is suitable for R

@bittersweet @bioDan
So is exel, matlab, mathematica, labview and 1000 other obscure academic data shuffling tools.

@Gregozor2121 there are many similar features because they all deal with data. But they vary tremendousely by the goal of the app and the way they go about managing and manipulating the data.

Its like saying all software is the same because it deals with memory, disk, and IO.