cs-guide

Some guidelines for computational scientists.

Introduction

Guidelines

• Learn a scripting language. E.g. Python, Ruby, Perl. As a fallback, at least learn some modern shell (e.g. zsh, Bash).

• Learn a programmable statistics package.

• Save everything in text files.

• use config files.

• Delete data you don't need.

• Learn the art of renaming data files and directories.

• off-the-shelf third party software, always generate complete config files in a script.

• Learn to use a command-line plotting program. E.g. gnuplot, R, mathplotlib

• Get a real programming editor.

• Automate everything

• Beware of the NIH (Not Invented Here) syndrome. FBR: this is maybe better describbed as doing orthogonal modifications to OTS software

Meta guidelines

Work smart (related to automate everything).

Conclusion

References

Bosco K. Ho's blog post "Notes to a young computational biologist" http://boscoh.com/protein/notes-to-a-young-computational-biologist.html

Article: Best Practices for Scientific Computing DOI: 10.1371/journal.pbio.1001745 http://www.plosbiology.org/article/info:doi/10.1371/journal.pbio.1001745

Book: The Art of UNIX Programming Author: Eric S. Raymond http://www.catb.org/esr/writings/taoup/

Book: the pragmatic programmer http://pragprog.com/the-pragmatic-programmer

How to do Research At the MIT AI Lab http://www.cs.indiana.edu/mit.research.how.to.html