I should also point out that there are other *excellent* steps in the "smart microscopy" + optogenetics direction, including recent efforts by the Pertz and Kapitein labs, that are reviewed here. Take a look! www.degruyterbrill.com/document/doi...

Our data suggested that differences in wave speed are sufficient to determine migration direction. And indeed, when we stimulate MDCK tissues with linear waves at different speeds, we observe a reversal of direction when wave speed increases!

If you look carefully, you'll notice something wild! Cells on the inside of the rotating-bar pattern generally follow the bars, twisting in a counterclockwise fashion. But cells on the outside go the opposite direction - moving in a *clockwise* direction! This effect is also clear in quantification.

OK, but how does this compare to a traveling wave of light, where cells have to respond to a macroscopic cue traveling across the tissue? Amazingly, we get migration in this case too but with some key differences.

OK, so what about cell migration? Can we optogenetically stimulate every individual cell in an epithelial tissue? How does this compare to signaling waves? PyCLM made it easy to segment all the cells in an MDCK OptoEGFR monolayers and give each cell its own tiny gradient. Voila!

I won't dig into Python code here, but crucially Harrison's tools are incredibly easy to use. You can "re-mix" any prior combination of experiments using simple text configuration files, and writing new experiments is straightforward. Here is an example of single-cell optogenetic feedback control!

Maybe surprisingly, the biggest challenge here was *software*! How can we automatically tailor light inputs to 100s of cells or deliver waves across a tissue? Harrison solved this problem beautifully by building a suite of tools for microscope control, cell segmentation, and optogenetic stimulation.

A long dream of my lab has been to interrogate tissue-scale movement with optogenetic receptor tyrosine kinases (like EGFR). With light, we can deliver local directional inputs to each cell in a tissue, or apply waves of light to a tissue at user-defined speeds.

Well, how tissues move naturally? One surprise is that tissue movement is often accompanied by long-range propagating waves of receptor tyrosine kinase activity (see work by Aoki, Di Talia, Matsuda, Ebisuya, & others). One prevailing notion is that these waves direct tissue movement.

Biologically, this work was motivated by a simple question: how can biochemical signals make a TISSUE move in a user-defined way? To move a 1000-cell tissue, should directional stimuli be delivered to each of the 1000 cells? Or can one take advantage of the collective?

New Toettchlab paper alert! Harrison Oatman just published his beautiful work combining software design for "smart" microscopy+optogenetics and some really beautiful work understanding EGFR-driven collective cell migration. www.cell.com/cell-systems...

Check out those figure making skills! She should do Illustrator tutorials.

There's lots more in the paper, but I just want to emphasize that an ideal switch: - Should be easily adapted to respond to other kinases. - Should easily "plug into" other target proteins. Both are true! For example, here is a "phospho-nanobody" that binds actin only when ERK phosphorylates it :)

Here's where Qinhao rolled up his sleeves and got to work, testing dozens of variants and optimizing every component to build a better phospho-switch. His final version basically looks as good as regular Gal4 when ERK is on, and has a 20x change in gene expression between ERK-on and off states!

So the idea: - Take an opto-Gal4 transcription factor we previously made by inserting the AsLOV2 switch - Swap out AsLOV2 for a FRET biosensor for our favorite kinase, ERK - Check if ERK activity now controls Gal4-induced gene expression! It worked, but honestly, not that well...

Qinhao had a stroke of insight: a classic kinase biosensor design (FRET biosensors) look just like an opto-switch! Both have N and C termini that are close together in one state and far apart in another. It is this change in N-to-C distance that "pulls" on the target protein to turn it on or off.

This problem looks a lot like one we face in optogenetics: Given a protein, can I make a light-switchable version of it? One way to do that is to fuse an "opto-switch" domain to a target protein. Light changes the conformation of the opto-switch, which tugs on the protein to turn it on or off!

I'd like to share a little bit of happy lab news in these chaotic times: a new preprint, driven by the brilliant Qinhao Cao! www.biorxiv.org/content/10.1... We address a big challenge in synbio: If you give me a protein "X", how can I give you a version of X whose activity is controlled by a kinase?

How do we expect science to be done around here if all the mugs are broken?