Exercises

[antagomir]

We should have some exercises about all major chapters / functionalities.

List systematically what is missing and create issue/issues.

At least:

• dbRDA exercises will require an update
• calculating beta diversity with altexp?

[RiboRings]

It would be nice to do the following in preparation for Radboud Summer School:

• give more detailed steps in subsetting exercise (see for example exercise above, which has nice numbering, instructions and hints to functions). This will be used in tutorial for data manipulation
• Improve exercise on subsetting by prevalence. For example, subset to the most prevalent taxa (17.3.3) (tip: subsetByPrevalentFeatures). for count assay; how many features remain? for relative abundance assay; how many features remain? This will be given as extra exercise in tutorial for data manipulation

[RiboRings]

@antagomir could we assign someone to this issue?
More ideas will probably follow as I continue writing the tutorials.

[antagomir]

Hi @Insaynoah and @AxelDgn could you have a look at this (and coordinate among yourselves)?

[RiboRings]

Another point that would be useful to improve for the upcoming summer school:

• Add step-by-step instructions in exercise 9.2 on heatmaps
• Create an additional exercise on heatmaps where some more advanced heatmap functionality is demonstrated

You can use this tutorial to get some ideas.

[antagomir]

The heatmap example could then also be included in the workflow.

[RiboRings]

Sorry, what workflow are you referring to?

[TuomasBorman]

This probably #501

[antagomir]

Ye.s

[TuomasBorman]

The exercises should be improved. Especially, we should check that they follow the latest mia.

Moreover, the example solutions are not tested during the build. This might lead to solutions that are not actually working. We should either remove the solutions completely or then make them runnable (eval: false --> output: false).

It would be nice to have example solutions, but on the other hand these exercises should be already solved based on the chapters. Keeping the solutions increases the maintenance burden. Also running them uses some additional computing resources.

If we decide to keep the solutions, we must check that the quality of them is good, currently, there are solutions that could be improved.

My opinion is that we could keep them.