pyrolite Presubmission Inquiry

[morganjwilliams]

Submitting Author: Morgan J. Williams (@morganjwilliams)
Package Name: pyrolite
One-Line Description of Package: A set of tools for getting the most from your geochemical data.
Repository Link (if existing): https://github.com/morganjwilliams/pyrolite

---

Description

• Include a brief paragraph describing what your package does:

pyrolite provides tools for munging, transforming and visualising geochemical data from common tabular formats. It enables you to recalculate and rescale whole-rock and mineral compositions, perform compositional statistics and create appropriate visualisations and also includes numerous specific utilities (e.g. a geological timescale).

Scope

• Please indicate which category or categories this package falls under:

  • Data retrieval
  • Data extraction
  • Data munging
  • Data deposition
  • Data visualization
  • Reproducibility
  • Geospatial
  • Education
  • Unsure/Other (explain below)

• Explain how the and why the package falls under these categories (briefly, 1-2 sentences). Please note any areas you are unsure of:

pyrolite leverages Pandas to enable import, munging and transformation of geochemical data from standard tabular formats, and matplotlib to facilitate common (and some less common) geochemical visualisations. One of the principal project aims is assisting to improve the reproducibility of geochemical research (especially for data-processing steps which often are overlooked or undocumented).

With regards to education, pyrolite is well suited to being incorporated into university-level geochemistry and petrology classes which wish to teach a little Python. The documentation is continually evolving, and more examples and tutorials will gradually be added. It isn't a principal aim of the project, however.

• Who is the target audience and what are scientific applications of this package?

pyrolite is targeted towards geochemists and geoscientists who use geochemical data (chemistry, mineralogy and relevant properties), especially those using lithogeochemistry. pyrolite has been developed principally to enable more reproducible data import, munging, transformation and visualization for geochemical data. In addition to this, pyrolite:

• Encourages better practices throughout these processes, including the use of compositional statistics (i.e. log-transforms).

• Implements some common geochemical models and methods to make these easily accessible and reusable (e.g. lattice strain models, orthogonal polynomial decomposition of Rare Earth Element patterns - 'lambdas').

• Contains a small database of rock forming minerals for normative calculations and looking up mineral formulae.

Extensions beyond the core package are also being developed for specific applications or interfaces (e.g. to alphaMELTS).

• Are there other Python packages that accomplish the same thing? If so, how does yours differ?

There is at least one other Python package which has some minor overlap for visualisations (GeoPyTool, which has a GUI-focused interface), but generally there are few open source Python packages for geochemistry (especially on PyPI). pyrolite provides some broader functionality (for both plotting and handling geochemistry) and is designed to be used from an editor or terminal and encourage geoscientists to further develop transferable Python skills. Where practical, the APIs for the tools on which it is built are exposed (e.g. pandas, matplotlib and sklearn).

• Any other questions or issues we should be aware of?:

P.S. Have feedback/comments about our review process? Leave a comment here

[lwasser]

thank you for this submission @morganjwilliams !! this looks like a great fit for pyopensci . i'll be in touch about next steps. i'm just catching up from AGU!! :)

[morganjwilliams]

Happy New Year @lwasser! I hope you've had a bit of time to relax and recover from AGU. Just following up on this one, let me know if there's anything to address before proceeding to a submission.

[lwasser]

hi @morganjwilliams !! my apologies for the delay. i'm catching up post holidays and beginning of the semester. We have a meeting this thursday but i suspect we will ask for a full submission after that meeting. you are welcome to come to pitch the package as well if you'd like!! it's at 11am mountain time. If not i'll just bring it up as an issue.

In the meantime if you could think of people who would be good to review this package, that would be wonderful.

[lwasser]

oh also @morganjwilliams HAPPY NEW YEAR to you as well!!

[morganjwilliams]

Thanks for the update, sounds good. I won't make the meeting this time around (2am my time), but I'll jot down some ideas for reviewers and start getting things ready for the full submission.

[lwasser]

@morganjwilliams we'd like to get this into our review process. can you please submit it?
Also if you can suggest any reviewers, i'd greatly appreciate it. if not we can post on twitter / discourse as well once you submit! many thanks for your patience. i got a bit behind over the holidays!!

[morganjwilliams]

Thanks @lwasser, I'm getting onto this now. It'll go into our internal publications system today, and should be able to be submitted shortly afterwards. I have an idea of one or two reviewers who might be interested, I'll ping them in the issue thread when I get the submission in.

[lwasser]

closing this given #20 thank you again @morganjwilliams !!