Reproducible Quantitative Methods
Lesson 7
Topics and Resources
-
Even more with R (writing simple software)
Review everything you did last week, and do some more! This is where things start getting very freeform and directed by your data- start really working towards the scientific questions your students asked of the data.
Depending on where you are with your data and analysis, we may want to introduce plotting with ggplot2- we can work on this in the seminar.
-
Getting a bit abstract
It looks like there are a couple opportunities coming up to present our work! Cool! There's the EEBB symposium on May 1, and then there's ESA in Portland this summer. If even one of your group members plans to attend either (or both!) of these events, I think it would be useful to everyone if we could put together posters- it's a line on the CV that is more immediate than publication, and it's a great opportunity to talk about the work and your experiences in the course. However, even if no one is attending anything, we're all going to need abstracts for our projects eventually- so let's see what we can get done on this on Tuesday. The deadline for ESA is SOON (the 23rd of February). The deadline for the EEBB symposium is farther away (Mar 31).
Exercises
- Loops, Conditionals and functions
And student directed analyses! Look! See what you find! This is the fun part :)
Discussion
Authorship on data, authorship in collaboration
Now that you're a little deeper into the course, you're ready for a more in-depth discussion about scientific authorship. The readings here are meant to present some extreme views and really provoke discussion. Please read these readings prior to class on Thursday.
Readings
Pubwication of software papers, and authorship on them. This is a blog post by Titus Brown, a professor of data-intensive biology at UC Davis. Titus’ group built a software product, khmer, out in the open, and invited contributions from the community. Long story short, he awarded authorship to everyone who contributed when he published a paper on the subject, no matter what the contribution was. This sparked a big debate in the community about what authorship really meant.
Data Sharing (an editorial from the New England Journal of Medicine).This editorial presents another view of authorship- namely, making the argument that data is essentially owned by those who collected it, and that no derivative works should be created without the data owner’s permission.
A Fundamental Difference of Opinion is blog post by me in response to the above editorial.
Questions
Why is there a discrepancy between the way credit is offered to data and software?
Are data and software research products?
What is the best way to credit data and software creators?