Forum:Code Sharing And Publications

Question

[Damian Kao]

Recent article on code sharing in publications:

http://www.sciencemag.org/content/336/6078/159

I agree with the article that scripts/software should be open sourced. But I also wonder how the journals will determine if a script is acceptable for publication. Are bioinformaticians going to have to spend extra time making scripts into a pipeline? Are all analytical scripts going to have to be curated and packaged? Do we run in to the danger of turning the biology source code landscape into a mess of spaghetti code by mandating everyone submitting their own version of even simple operations?

Answer 1

[Istvan Albert]

The publication that you link to suggests mandating source code release for published data analyses - and that's a good thing.

I do think that bioinformaticians should already be spending substantial amount of effort on properly testing, documenting and packaging their code, this makes it more reliable.