List of lessons

[m-rivera]

Hi All,

We thought it would be good to start a project map listing the lessons that we need to include in the website. Here is a first go, inspired by the "official" Python tutorial:

• Running a Python script or Notebook (both already exist)
• Variables, data types, and operators
• Conditionals
• Loops
• File handling
• Visualising with matplotlib
• Numerical operations with numpy

I'd welcome any thoughts or edits!

[maddicoat]

The title is python in chemistry, so:

• ASE (https://wiki.fysik.dtu.dk/ase/index.html)
• RDKit (https://www.rdkit.org/)

Here is my version, which has now been running successfully for 6 years: https://maddicoat.github.io/CHEM10121/

[m-rivera]

Hi Matt, thank you for sharing that! I'm always a fan of the Software Carpentry style guide, and the curriculum you've set up at NTU looks very solid.

I presume that it's chunked in lessons of a consistent length?

I'd be very keen for us to add modelling and chemoinformatics lessons as standalone one-offs. Our finding from last year's meeting was that many colleagues didn't come from a modelling background and wanted to use Python (in their teaching and research) for things like experimental data analysis. I concede that RDKit is adjacent to that, but not sure it will be essential for everyone.

[maddicoat]

Hi Miguel, I agree and I would add pandas and seaborn in that direction.

For now, the module that I fit in doesn't have the time for me to do that. I teach this at first year level, where for many folks it's their first programming ever. Scaffolding everything towards the ASE permits an 'easy sell' that this is indeed python for chemistry. Having to teach / sell statistics + python simultaneously sounds challenging... :-/ I make sure to mention that in one term I can't teach "all of python" and that the aim is to teach enough of the basics such that a student could look up a new / unfamiliar library and work it out using the time-honoured trial-and-error approach.

I teach it in hour long slots. Plotting and functions (Ep 5 and 6) can fit together. I take some extra time at the end of Ep 1, to extend the "famous_chemist" dictionary into a list-of-dictionaries - students can choose more chemists on their own - I point them towards two ACS resources : https://cen.acs.org/people/lgbtq-scientist-chemist-history/99/web/2021/06 and https://cen.acs.org/people/profiles/Six-black-chemists-should-know/97/web/2019/02?sc=231026_mostread_eng_cen.

Beyond the workbook, there is one lecture on pseudocode and one lecture based around the assessment (primarily how to translate the assessment requirements into code - e.g. "rotate this molecule through a full 360 degrees in 5 degree increments == for this_angle in range(0,365,5):"

[m-rivera]

After a chat with Alan, we've come up with the following list:

1. Variables data types
2. Mathematical operators
3. Comparison, conditions, and indentation
4. For loops
5. While loops
6. Reading files
7. Writing files
8. Defining functions and scope
9. Matplotlib basics (plot, scatter, fig ax)
10. Understanding documentation (for matplotlib)
11. Numpy basics (arrays, operations)
12. Numpy functions (axis=, argmin max, linalg, stdev)
13. Pandas (maybe multiple lessons)
14. Optional extra libraries (e.g. ASE, RDKit, seaborn, scipy)
15. Optional case studies (complete exercises that involve multiple skills and have applications to Chemistry)

[ELRIC1982]

Where may I find these actual lessons? New to the site, but not to chem or python. This looks amazing already.

[m-rivera]

Hi :) !

The lectures still need to be proofread and merged, but you can check them out on the open pull request here:
pythoninchemistry/pythoninchemistry.github.io#46