You should try BioVinci, a free and beginner-friendly statistics web application can produce publication ready figures in minutes. Just drag and drop your data.

You could try the some of examples here: https://vinci.bioturing.com/panel/workset/build/create-grouped-violin-plot

https://vinci.bioturing.com/panel/workset/build/hierarchical-clustering

https://vinci.bioturing.com/panel/workset/build/tsne

R

Jochen, I developed a simple method that may result useful for your question. I do not mean it is the only possible one, it simply works fine, provided that you add a few premises from yous experience and empirical results. OK, that is all. emilio

R allows you to do every data analysis and graphs with a very high quality.

As usuaul, there is no "best software". From my point of view, Matlab can be a good tool to explore data with possible import features.

I would recommend R, because the results require fewer formatting and immediately publishable. It is also good at multidimensional graphs.

I also recommend R (www.r-project.org). You can find many libraries for different statistical models

R of course!

I would suggest "Statistica" as a more user-friendly software. However, "R" performs all analysis you need.

The program selected depends on how complicated the analyses and graphs are and how easy the program is to use. Currently, I am developing a general univariate-multivariate GLM program which can analyze any data. It includes appropriate 2 dimensional graphs and tables that are ready to be placed in a paper or used in a presentation. It is for those who wish to make decisions from their data (rather than looking at multiple statistics).

UniMult.com.

The choice of software will ultimately depend on the nature of the dataset, the complexity of the analysis needed. Mainstream tools have comparative strengths in certain aspects. None really would be a classic "Jack of all trades".

One would go with the a software that covers the most critical aspect of what needs to be achieved; whether it is the analysis or the generation of appropriate graphs representing the data. With the plethora of tools out there and the compatibility of data formats, most likely a hybrid of sorts will give the best outputs across.

@Andrew: in my opinion (and my experience) R is a "Jack of all trades". Even when I need to go outside of R for some tasks I still find that the optimal solution makes use of R. For instance, for some tough computational jobs R code is very inefficient while C++ allows one to write extremely fast code. In that case I use the "Rcpp" package in order to do the data preparation in R, the specialized computation in C++, and finally display and interact with the results in R again. The Rcpp interface is incredibly elegant, almost painless. Similarly, I write reports or scientific papers ultimately in LaTeX but again there are wonderful interfaces such as Sweave and more recently knitr. Precisely because R is open source and widely used, this kind of solution has been implemented elegantly for all kinds of "interface" problems.

You should try BioVinci, a free and beginner-friendly statistics web application can produce publication ready figures in minutes. Just drag and drop your data.

You could try the some of examples here: https://vinci.bioturing.com/panel/workset/build/create-grouped-violin-plot

https://vinci.bioturing.com/panel/workset/build/hierarchical-clustering

https://vinci.bioturing.com/panel/workset/build/tsne

If you code well, R is great. I also recommend BioVinci, quick and easy for non-programmers.

State would be my choice, but requires a lot of coding.

R, for sure ! If you need any examples of ready-to-use scripts, take a look at my repository and feel free to reuse them under Code section:

https://www.researchgate.net/profile/Polina_Lemenkova/contributions

the same scripts (and more), + with resulting image outputs as pdf you can find on my github (40 ready-to-use scripts for various kinds of statistical analysis):

https://github.com/paulinelemenkova?tab=repositories

After installing R you need to install various libraries (depending on what exactly you need to draw), but some basic functions and visualization tools are already built-in in the core version of R.