Tweak print method to show altExpNames

[Biomiha]

Add printing of altExpNames if they exist or NULL if there is nothing.

[stemangiola]

Thanks! Well done.

For some reason, you have conflicts to solve (see above). Maybe recent PR changes have the same lines you change. Please have a look and solve the conflicts.

Also, I see you have two branches, but you are doing the PR just for master. Not sure if this is intended.

[stemangiola]

Also, please add a unit test, add a dummy "alternative experiment", and verify that the print looks as expected.

[stemangiola]

Thanks.

I bit more elegant would be

• Replace altExpNames with Assays_alternative
• Conditionally add "| altExpNames=%s" if alternative experiment exists

[Biomiha]

Fair point. I had considered adding the information on a conditional basis but personally I prefer to know even if it's empty. The standard SCE print method outputs altExpNames(0): when empty and altExpNames(1): Antibody Capture when it is populated (Antibody Capture being the name of the alternative experiment and can be any string). If you feel you would rather keep it only when it is non-NULL I can do that quite easily.
In relation to the naming, it actually refers to the name of the experiment rather than the names of the assays, i.e. if it exists you will still have counts and normcounts or other as the assay names within the altExp slot. How about saying Other_experiments=?
I will also add the unit tests and other missing bits whilst I am at it.

Thanks.

[stemangiola]

I see.

let's try to generalise the concept of experiment and assay, still hinting at the non-modular nature of singlecellexperiment.

So I am in favour of printing the experiments within the object, if they are 1,2,3, independently whether they are base or alternative.

having said that, there are a lot of ways to go about that. One possibility is to do

Assays= base:counts,logcounts; Antibody capture:counts,logcounts; ADT:counts

Other ideas are welcome!

[stemangiola]

@Biomiha, I realised that the sentence "OK, the goal here is to provide an interface that promotes modularity. " sounded very bad in writing; the tone I was thinking was completely different from how you would read it!

I used a poor form to give the broad indication that "the tidy transcriptomics thrives to giving modularity (which sometimes diverges from the Bioc framework)."

I hope you did not take it personally!

[Biomiha]

No dude, absolutely not. I work in biotech 😄 .
I agree with what you've said. Am working on implementation, just been very busy lately.