developing a package for the Tree of Sex database

[Alex Harkess]

Hi everyone,

My name is Alex Harkess, and I'm a graduate student with Jim Leebens-Mack at UGA. I'm really interested in learning how to develop packages for R, especially in the context of rOpenSci. I'm hoping to jump right in and make one of my ideas a reality -- creating an interface to work with the Tree of Sex database (http://treeofsex.org/). Briefly, this database is a huge aggregation of vertebrate, invertebrate, and plant sexual system information. Unfortunately, the authors didn't do much in the way of making the data usable outside of a spreadsheet they put online. Speaking specifically about plants, it would be great to be able to parse this database and attach information about sexual systems (dioecy/hermaphroditism, ploidy, self-compatability/incompatibility, etc) to the Open Tree of Life treetips, allowing people to extract clades of interest and, for example, perform their own ancestral state reconstructions. Since David and others are working on an API for rOpenSci, I gather that it makes a lot of sense to use rOpenSci as a intermediate for this kind of species information.

I have an intermediate skill level in unix, perl, python, and R, but am just starting to learn how to really develop usable packages meant for sharing. A few people from rOpenSci have reached out to me already to get this project started (https://github.com/alexharkess/tosr), and I am already learning quite a bit about how to properly develop a package. At the same time, I'm hashing out some of my ideas about practical use cases for this data, not just for myself (plant-centric), but for the vertebrate/invertebrate crowd as well.

I wanted to reach out to you all to let you know about this idea, and I'd love to work together and learn from people here to develop my first (hopefully) useful package that could become part of rOpenSci, too.

Thanks,
Alex

[David Winter]

Hi Alex and all,

Looking forward to working with you on this package and other R
projects for the Tree of Life meeting.

It occurs to me that package development in R can be a slightly
bewildering topic for someone transitioning from "run my analyses in
R" to "software development in R". Documentation, vignettes, roxygen,
testing, devtools and continuous integration, version control and
github all take some learning.

I wonder if it's worth while making an "intro to package development"
page of community section of the website? It could mostly consist of
links to existing tutorials and the like, but a clear explanation of
how to get started on package development might be helpful?

If people think it's a good idea Alex and I can start work on that
page as we work on the new package, and point others to a draft page
as we go?

Cheers,
David

> --
> You received this message because you are subscribed to the Google Groups
> "ropensci-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to ropensci-discu...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.



--
David Winter
Postdoctoral Research Associate
Center for Evolutionary Medicine and Informatics
The Biodesign Institute
Arizona State University

ph: +1 480 519 5113
w: www.david-winter.info
lab: http://cartwrig.ht/lab/
blog: sciblogs.co.nz/the-atavism

[Scott Chamberlain]

Alex, Super excited about the pkg for the tree of sex database!  Ping us if you need any help. (this mailing list, mention us in github issues/email/twitter/etc. )

Best, Scott

> email to ropensci-discuss+unsubscribe@googlegroups.com.

[Alex Harkess]

"It occurs to me that package development in R can be a slightly bewildering topic for someone transitioning from "run my analyses in R" to "software development in R"."   Definitely where I'm at! I love using R, and have written my fair share of one-off scripts to mung/summarize/plot, but reading others' developed code is a bit daunting.

I'm on board with helping to write a "how-to" page (especially from the perspective of my learning-by-doing). I imagine I'm probably a typical use case...someone who can write a little informatics code, but needs some hand-holding to move up to developing.

Thanks David et al!

[Jeffrey Hollister]

David and Alex,

Just to get you started, these are the links I relied on when I first started developing my own package.

1.) Hadley's page/book on package development is how I moved into developing packages.  I think it is pitched at the perfect level for those jumping into package development. http://r-pkgs.had.co.nz/

2.) The other thing that helped me was using the RStudio tools.  They too have a good page on this: https://support.rstudio.com/hc/en-us/sections/200130627-Package-Development

3.) Hilary Parker has a blog post on the topic that has been getting some recent and well deserved love on Twitter: http://hilaryparker.com/2014/04/29/writing-an-r-package-from-scratch/

4.) Jeff Leek's Lab page on github is a good resource: https://github.com/jtleek/rpackages

5.) Dense, difficult, but official! http://cran.r-project.org/doc/manuals/r-release/R-exts.html

Cheers,

Jeff