Cytometry in R - Week #13 - Spectral Similarities

[UMGCCCFCSR]

Hello everyone, we are now at the thirteenth week of Cytometry in R course, where we will start comparing fluorescent signatures by various metrics! For previous course emails, see here.

Community Feedback Survey # 2

Since this is our first year offering the course, we will sporadically ask you to fill out a feedback survey to evaluate what is going well, what areas can be improved on. Thanks to everyone who has so far completed the community feedback survey #2. So far, preference seems to be toward weekly consistency (even if it means breaking up material across multiple weeks). The survey will remain open for another week and can be found here.

If you had a specific question or suggestion (where to find such-and-such), I will be posting answers/thoughts in the Discussion forum on Wednesday. If your ggplot2 - ggcyto - flowGate - Luciernaga package installs are having issues and you want help with bug fixes, open a Discussion post! Run quarto check in your terminal, screenshot the output, and attach it.

Week # 13

This week for the primary course material, we will utilize the spectral signature matrices we have isolated over the last two weeks from our unmixing controls, and evaluate different ways of comparing different fluorescent signatures with each other. In the process, we will gain better understanding of various commonly used metrics, including similarity (cosine), complexity (condition number reference matrix/kappa), and unmixing-dependent spreading (collinearity). Additionally, for the bonus content, we will take a look at spillover-spreading and staining index reduction matrices. You can find this week's primary course material here.

For those attending online:

• Tuesday, July 28th 2200 EST (Wednesday 0200 GMT+0)
• Wednesday, July 29th 1530 EST (Wednesday 1930 GMT+0)
• Thursday, July 30th 1000 EST (Thursday 1400 GMT+0)

All recordings will be available immediately afterwards on YouTube.

For those attending in-person

Still waiting on CC to get back for new material, if we want to review I am willing to meet on

• Thursday, July 30th 1600 EST (Thursday 2000 GMT+0)

Next up

On the docket for next week, we will tackle primary course content in the form of "Week 14 - Unmixing", where we will take the signature matrices we have extracted/validated over the past three weeks and use them to unmix the full-stained samples for this dataset.

Two weeks from now I will be in Seattle, WA for the Bioconductor conference. My plan is to schedule recordings related to the unmixing primary content to release throughout that week for anyone wanting to maintain learning consistency.

Best Wishes-

David