I agree. I think it would be good to know "R" and "Python". I don't think you can say you are a full computer scientist if you don't know the C languages. However, there are a lot of tools available in "R" and "Python" and if you want to get really fancy, you can pipe "R" commands into "Python" and vice-versa, although there are some limitations with that. I think that if you want to consider yourself a bioinformatician, you need to know how to write your own functions, algorithms, and objected oriented code. It's easier to write scripts yes, but scripts are specialized pieces of code and they don't have much scope.
In terms of phylogenetic tree construction, there is a nice software called "MEGA" which has very advanced methods of phylogenetic tree construction. It would be good to understand how those methods work to judge the probabilities of each node. However, remember that your observations are only as good as the quality of data that you input.
As for analyzing protein structure, you can use "Pymol" to analyze protein structure from Protein Data Bank files, and you can also write scripts in it. I used to have more confidence in the deposited protein structures, but now I'm a crystallographer so I see some imperfections.
The choice of a language is a subsidiary question. Two main questions must be answered before: (i) what is the problem that I must solve? and (ii) what is the best informatics described in the bibliography to explore similar problem? Good luck, Christian
Christian Cibert, so effectively the algorithm is most important. Some programming languages might supply better scalability, user interfacing, or speed of implementation.