Thread by @_rdgao

holy shitballs - I mean brainballs - our oscillating organoid paper finally, finally got published last week.

I wrote about the entire (5-year) process and explained the science here:

rdgao.com/oscillating-or…

It's an insider's look into this whole thing. I think it's hilarious.

The paper: cell.com/cell-stem-cell…

The article got featured in the New York Times last week in a piece by @carlzimmer, and I like it a lot, especially its balanced take on what's shown and, more importantly, what's not shown:

nytimes.com/2019/08/29/sci…

ok now my obligatory tweet thread:

Human stem cell-derived brain organoids have been successful for modeling development and disorders from a gene expression and structural point of view.

But complex spontaneous activity had not been shown, which is important in-vivo.

In this paper, we show exactly that:

Spontaneous and OSCILLATORY (~3Hz) activity developed in the organoids starting from 4-months, and we can record consistent population spiking and local field potential signatures.

(of course, we're oscillation nerds, this is what we do.)

The oscillations actually disperse by 8 months, and these population events, in general, become more random in space and time.

Meanwhile, we can track cell type emergence over time as well, and it seems GABAergic cells and glial cells have important roles here.

The kicker of this study, of course, is that we compare LFP in the organoids to preterm infant EEGs.

Preterm baby EEGs (from 25-38 weeks old) exhibit this sparse on-off kind of dynamics, which is not seen in adult EEGs, but looked very similar to what we see in the organoids.

We did a bunch of analyses to compare them, but the highlight is that we trained a simple (regularized) linear regression model on JUST the baby EEG features, then used the model to predict the organoids' culture age.

Surprisingly, it did...alright. Even though variance is really high across the organoids, on average, prediction was close to actual culture time.

This is a complete mindfuck.

Think about it, what's absolute time to these organoids that have no knowledge of the external world?

So my interpretation is that maybe there is some genetically preprogrammed developmental trajectory, not only for cell differentiation and migration, but also spontaneous electrical activity.

The interpretation is NOT "we grew baby brains in a dish."

And if you're wondering about all the science fiction possibilities, or the ethical concerns, go read the blog post. Those things need way more space to unpack than 140-chars.

I also cuss a lot in the blog post because it made me hate some aspects of peer review.

I will end here by saying that this has been...a journey.

Obviously, thanks to everyone involved: Alysson Muotri, @bradleyvoytek, & in particular, Pri Negraes & @trujillo_cleber, who owes me A LOT of beer.

It's been fun, I'll do it again, but hopefully it doesn't take 5 years.