iiinova — where your digital self keeps evolving

Yes, Hunt et al’s ‘athorny‘ are part of the deep PN population! And their beautiful in vivo recordings will be directly applicable for understanding coding functions

Thanks a lot!! and also for the AMAZING open dataset! It’s really an incredible resource

010

Using patient tissue, we find distinct properties of deep and superficial cells in humans, with species-conserved functional hallmarks. Intriguingly, deep CA3 cells were shown to have a GWAS link to human intelligence..! Perhaps these 'Shepherds of CA3' play a crucial role in cognitive function..

Image of a patch-clamp recorded human hippocampal slice, with firing profiles of recorded neurons (centre) showing higher burst firing in deep cells. An image from the GWAS analysis of Thompson et al. Nature Neuroscience 2025 (reproduced under CC BY-NC-ND 4.0; right) highlights the association between deep CA3 cells (CA3.2) and 'Education years' and 'Intelligence'.

180

And this isn't just a mouse thing! Using more fantastic #opendata, we can see that the two CA3 cell types are conserved in both pigs and humans! 🐖⛹️‍♀️

Transcriptomic analysis of pig (left) and human (right) hippocampal data shows a subpopulation of CA3 cells specifically expressing St18 in both species (arrowhead), identifying the conserved deep CA3 cell type.

2101

But there's a big surprise with projections! Sup CA3 cells form the textbook associative network we know and love. But not deep cells. They form hemisphere-specific modules that appear built to regulate the superficial network through excitation and FF inhibition. A new hippocampal circuit module! 🦚

AAV labelling of CA3 cell types (red, superficial; blue, deep) shows their different projection patterns. Superficial cells project to CA3 and CA1 in both ipsilateral and contralateral hemispheres. Deep cells show a highly specific unilateral projection, mainly within CA3, at the base of stratum radiatum. A circuit schematic is shown on the right.

1110

We mapped the differentiation of these cells in development. The two cell types are not born strictly sequentially, but we see preferentially early generation of deep cells.. following general rules of the cortex!

Neurons were birthdated by BrdU labelling (green) in St18-Cre mice (blue) at three developmental timepoints. Labelling is graphed (right) showing decreasing St18/BrdU co-labelling across development.

170

We made a mouse line for genetic access to these two cells. Patch-clamp recordings show that deep St18+ PNs unify the bursting, long-shank, athorny, and sparsely thorny neurons reported across the last 40 years!

Neuronal labelling in St18-Cre mice (blue cells) shows specific labelling of deep CA3 neurons (upper right). Properties of patch clamp recorded cells (yellow, lower left) are plotted (lower right), with higher burst firing, longer primary dendrites, and less thorny excrescences in deep CA3 cells.

1132

These cell types are deep and superficial neurons, with the St18 gene marking deep CA3 cells. CA1 has much better studied heterogeneity, but in contrast to CA3, deep and superficial CA1 cells behave more like a gradient than distinct cell types. CA3 has uniquely strong PN cell type identity!

Allen Brain Cell Atlas spatial transcriptomic data shows deep and superficial locations of the cell types in CA3 (upper left), with St18 gene expression showing only in the deep cluster (upper right). UMAP and Sankey analysis in CA1 shows that deep and superficial CA1 neurons form a gradient rather than distinct cell types (lower).

190

Heterogeneity is everywhere in the brain, but making sense of it isn't easy. To understand CA3 neuron heterogeneity, we started from scratch using the amazing #openscience resources from the @alleninstitute.org. Surprisingly, reanalyzing CA3 showed not one, but two clear pyramidal neuron types!

UMAP depiction of CA3 neuron transcriptomic data reanalysed from the Allen Brain Cell Atlas showing two cell types (red and blue). Sankey diagram shows cluster stability across resolutions.

160

Jake Watson @jakefwatson.bsky.social · 10/07/2026

We have a new preprint! By studying neuronal heterogeneity in CA3, we found a weird but powerful new feature in the hippocampal circuitry! 🧠 #neuroskyence 🧵.. Morse-Mora et al. | Hippocampal CA3 forms a two-layer network of molecularly distinct cell types in mice and humans doi.org/10.64898/202...

Mouse hippocampal slice with deep cells labelled blue, patch-clamp recorded cells in yellow, and deep CA3 projections labelled purple

412236

But also - substantial in humans in our samples! Plenty of work to do yet even to understand the hippocampus. #backtotherig

100

Jake Watson @jakefwatson.bsky.social · 10/07/2026

Thanks Andrew!! Not quite non thorny, but certainly less thorns and encompassimg previously reported athorny and sparsely thorny cells. A very weird module on the hippocampal circuitry!

100

Jake Watson @jakefwatson.bsky.social · 07/07/2026

And it's a nice time to mention: from September, I'll be setting up human circuit research with Hospital del Mar in Barcelona. We want to use the amazing opportunity of surgical samples to explore how our own brain works. If you want to discuss ideas/collaborations/anything, lets chat at #FENS2026!!

Photo of Barcelona seafront showing Hospital del Mar and the PRBB

2160

Jake Watson @jakefwatson.bsky.social · 06/07/2026

And we have exciting new data on CA3 circuit architecture across species. 🐁🐖⛹️‍♀️ The hippocampus has another surprising twist for us! If you want to hear more, come to Rebecca Morse-Mora's poster (PS02-07PM-553) on Tuesday afternoon, 14:00-17:30

Patch clamp recorded CA3 pyramidal neurons filled with biocytin showing as white/gold on a black background.

050

Jake Watson @jakefwatson.bsky.social · 06/07/2026

If you are at #FENS2026, we have a symposium on human circuits - from single cells and synapses to intact circuit function! It's not easy to investigate how our own brain works, but we will present recent work showing that it is possible at many scales. Wednesday afternoon at 15:40 - Hall E.

FENS symposium advert stating: Symposium S29 - Wednesday 8th July 2026 15:40 - 17:10. Hall E

Untangling the human brain:
characterising the properties of human cells, synapses and circuits

Speakers:
Kimberly Siletti (Utrecht, Netherlands)
Christiaan PJ de Kock (Amsterdam, Netherlands)
Jake F Watson (Klosterneuburg, Austria)
Anna Maslarova (New York, USA)

2112

Jake Watson @jakefwatson.bsky.social · 08/05/2026

Happy birthday to the greatest Ambassador for Nature that we have. 100 years of teaching us the best reason to protect the natural world - it is incredible and beautiful! www.theguardian.com/tv-and-radio...

theguardian.com

‘A true pinch-me moment’: memories of David Attenborough as he turns 100

Stories of chance encounters at home and abroad and the influence of his programmes around the world

RSPB @rspb.bsky.social · 10/04/2026

Reposted by Jake Watson

To reduce disease, we all need to make some urgent but simple changes to how we feed our garden birds. See our website for the latest guidance: brnw.ch/21x1t4w

A Robin perched on a branch, its beak full of food and other material.

16652325

Reposted by Jake Watson

Andrea Navas-Olive @acnavasolive.bsky.social · 08/04/2026

If you are interested in systems neuroscience, learning about data analysis, or understanding how the conceptualization of analyses can influence results, join us for the Brainhack satellite event at #FENS2026! More info at: pre-fens-brainhack.github.io/brainhack2026/

02312

Reposted by Jake Watson

Eoghan Daltun 🌍 @irishrainforest.bsky.social · 16/03/2026

"The results are astonishingly clear. On rewilded land, biodiversity surged across the board, with the number of bird species up 261%. The variety of bumblebee and butterfly species more than doubled, their abundance increased over 10x." The cure for nature loss exists. share.google/w30RUdw1902W...

share.google

Rewilding linked to dramatic increases in birds across Scotland

New research across Scotland's Northwoods Rewilding Network has revealed striking biodiversity gains on rewilded land, with bird diversity increasing by more than 260% and pollinating insects rising m...

13764344

Reposted by Jake Watson

Rick Betzel @richardfbetzel.bsky.social · 15/12/2025

🧠🪰 The adult Drosophila brain connectome now gives us a complete wiring diagram of ~140k neurons. But a wiring diagram alone isn’t understanding. How is this massive network organized? Our paper tackles that question by mapping community structure across the entire fly brain. 1/

24325

Jake Watson @jakefwatson.bsky.social · 13/12/2025

Great to see it out! Congrats!

020

Reposted by Jake Watson

Yangfan Peng @yangfanpeng.bsky.social · 13/12/2025

🥳Excited to share our latest human multipatch paper, now out in @natneuro.nature.com 🧠 We studied the cellular and synaptic physiology of human L2–3 pyramidal neurons and identified subtype-specific local connectivity rules across individuals. www.nature.com/articles/s41... Join us: penglab.de

46725

Jake Watson @jakefwatson.bsky.social · 29/09/2025

And you'll find the preprint explainer here:

Jake Watson @jakefwatson.bsky.social · 29/09/2025

Our paper on 'mini analysis' is now published in @jphysiol.bsky.social If you are recording mPSCs or sPSCs, I hope this helps with analysis. Interpretation of these datasets is not as easy as it seems..! Happy to discuss if you are interested physoc.onlinelibrary.wiley.com/doi/10.1113/...

physoc.onlinelibrary.wiley.com

‘Mini analysis’ misrepresents changes in synaptic properties due to incomplete event detection

Abstract figure legend Summary of the study where simulated recordings (left) were used to characterise the effect of incomplete detection on mini (mPSC) analysis. Recording noise levels (red) determ...

1249

Reposted by Jake Watson

beatrizherguedas.bsky.social @beatrizherguedas.bsky.social · 15/09/2025

Extremely happy to share our lab’s first paper now published in @NatureSMB! We determined the cryo-EM structures of GluA4, both alone and in complex with TARP2, capturing the receptor in different functional states.

143

Reposted by Jake Watson

Timothy O'Leary @timothyoleary.bsky.social · 04/09/2025

Finding correlates of the same signal in different brain areas is not evidence against specialisation, it is evidence for interconnection. When I flush my toilet, the level in the tank and the flow rate in the supply pipe become perfectly correlated. Yet these toilet areas have distinct functions.

710421

Reposted by Jake Watson

mojtabart.bsky.social @mojtabart.bsky.social · 20/08/2025

Are you passionate about advancing cutting-edge imaging technologies, and combine expansion microscopy with the power of AI@HHMI to unlock new biological discoveries? If you’re ready for a creative,and impactful journey, we’d love for you to apply! hhmi.wd1.myworkdayjobs.com/en-US/Extern...

hhmi.wd1.myworkdayjobs.com

Research Specialist - Expansion Microscopy

Primary Work Address: 19700 Helix Drive, Ashburn, VA, 20147 Current HHMI Employees, click here to apply via your Workday account. The Howard Hughes Medical Institute’s Janelia Research Campus is a pio...

001

Jake Watson @jakefwatson.bsky.social · 04/08/2025

Hi Pablo - if you still need the info, this pattern is very nice staining of the mossy fibre tract (DG output). Your protein seems to be specific for granule cell axons!

110

There is a lot more on cell and synaptic properties in the full paper, but this data adds to the emerging picture of heterogeneous excitatory neurons... many interesting cell and circuit features can be revealed only by diving into the (often complex and confusing) heterogeneity of the brain!

Inhibitory wiring also had a surprise. Spontaneous GABAergic input was strikingly coincident on PNs of the same subclass, suggesting specific wiring of interneurons into each PN type - for layer-specific control! @zhihaozheng.bsky.social has now visualized this wiring in his beautiful connectomics

Composite figure showing (upper left), spontaneous PSC recording with correlated input into specific cells (purple). Cross-correlograms (lower) showed specific coincident input at t=0 onto pairs of PNs of the same subclass. Quantification (upper right) confirmed this effect, which could be blocked with the GABAAR antagonist GABAzine (purple).

120

Deep and superficial PNs had a surprise in their recurrent connectivity: superficial PNs connect broadly, but deep PNs show little local connectivity back to superficial cells. This suggests that deep PNs form their own subnetwork, perhaps 'associating the associations' made in superficial CA3..

Composite figure showing staining (lower) and schematic (upper left) of a multicellular recording containing two PN subtypes (superficial, red; and deep, blue). Right: Connectivity measurements within and between subtypes show similar connectivity from superficial cells to both other superficial cells and deep cells, whereas deep cells preferentially connect to other deep cells.

110

Across the brain, PNs show a lot of heterogeneity. In CA3 it’s clearest on the deep-superficial axis, and we see 'deep bursting' PNs that have been shown before. From hundreds of recorded cells, these are clearly a distinct subclass in our hands, but are not strictly 'athorny' as has been suggested.

Collection of figures showing (upper left), multicellular patch-clamp recording pipettes in tissue, (lower left) example superficial (red) and deep (blue) pyramidal neuron stainings, with indicated thorns (arrowhead), and (right) graph of burst firing frequency for CA3 pyramidal neurons showing segregation of two subclasses.

110