Action Potentials for March
1: A big connectomics study of Drosophila larva was published and made a splash on social media. It was first published on Biorxiv and like many papers, it’s available there open access. You can count on me to cover this kind of thing. Let’s dive in:
a: Previously, synaptic-resolution connectomes from volume EM had only been mapped for three organisms: C. elegans, Ciona intestinalis larva, and Platynereis dumerilii larva. That makes this the fourth.
b: They reconstructed 3,013 brain neurons with about 544,000 synaptic sites, with over 99% of neurons reconstructed to completion and the majority of annotated synapses — 75% — connected to a neuron. So to be pedantic, it wasn’t a completely full connectome, but it’s still the biggest to date.
c: Cajal’s law of dynamic polarization continues to be overturned, as some axons were found to contain postsynaptic sites and some dendrites to contain presynaptic ones. But it’s still a useful approximation. Axo-dendritic (a-d) connections were more likely to be stronger, as proxied by having a higher fraction of edges with more synapses between them.
d: Strong edges with 5 or more synapses between the connected neurons contain 54% of total synaptic sites, confirming the general principle in neuroscience that “a skeleton of stronger connections is immersed in a sea of weaker ones”.
e: They were able to cluster neuron types by their synaptic connectivity. Neurons within the same class tended to share the same NBLAST score, which seems like a nice way of quantifying the similarity of morphology between neurons.
f: They mapped all of the sensory pathways in the Drosophila brain, which is really cool to have all at once. They were able to classify input neurons based on their known sensory modalities, such as olfactory, gustatory, thermosensory, visual, gut state, respiratory state, mechanosensory, nociceptive, and proprioceptive. Most neurons in the brain are integrative (88%), which means that they ultimately receive neural information from more than one of these sensory modalities.
g: Dopaminergic neurons (DANs) had highly recurrent connections, with almost all of them being reciprocally connected with more than 50% of their partners, via multi-hop connections. They also all integrated signals from all sensory modalities. This organization potentially enables high-dimensional feedback in dopaminergic neurons for encoding distinct features, engaging in parallel computation, and contributing to working memory.
2: Theoretical study finds that simulated neural networks generated based on a small number of rules accurately replicate a wide range of properties in the connectomes of zebrafish olfactory bulb, mouse visual cortex, and C. elegans, including individual synapses and their strength, synaptic indegree and outdegree distributions, and sub-network motif frequency.
3: Konrad Kording tweeted a link to a Google Doc research proposal called “The complete neuronal input-output functions project”. It aims to comprehensively measure the Input-Output Functions (IOFs) of all 302 neurons in C. elegans, which characterize how neurons integrate and respond to inputs. No complete measurement of IOFs has been made in any animal, making this an ambitious project. It proposes the use of neuronal stimulation, electrophysiological recording, and molecular mapping to build models integrating all three of these levels.
Why does this matter? A key step needed to make connectomics data more useful is to be able to predict cellular input-output functions. Being able to do this even on one of the smallest organisms would be a huge advance for neuroscience.
4: A two-day workshop on whole brain emulation will be hosted by Anders Sandberg and Allison Duettmann at a venue outside of Oxford in May 2023.
5: Using a novel optogenetic pacemaker, a new study by the Deisseroth lab finds that increasing the heart rate can directly boost anxiety-like behavior in mice, and that this is mediated by the posterior insular cortex via bottom-up cardiac interoceptive processing.
6: New research suggests that the declining dementia rates of 20 to 25% in the past 3 decades are primarily due to better cerebrovascular disease — ie less atherosclerosis and arteriosclerosis.
7: Christin Glorioso on why life extension research is a good thing.
8: A list of active anti-aging trials.
9: The regular use of osmotic laxatives (like Miralax) is associated with an increased risk of all-cause dementia (HR 1.6) and vascular dementia (HR 2.0) in the UK biobank. The authors speculate this might be due to changes in the microbiome. I think this study needs to be corroborated before we put too much stock into it. It's associational and the causality is unclear. Still, it’s an association that I wouldn’t have predicted.
10: Decreased myelin proteins (MAG, PLP1) are found in the brains of people exposed to repetitive head injury.
11: New RCT (n = 150) finds that adding simvastatin was not helpful for improving depression scores when compared to placebo in treatment-resistant depression.
12: Even though we don’t hibernate, it seems that humans need more sleep in the winter months and often feel especially tired in February and March. Relatable.
13: New study uses MRI data to analyze the maturation process of association fibers, finding that the latency of rapid event-related potentials (specifically, the N1 component) decreases with age. As a result, transmission speed increases twofold from childhood to adulthood, indicating that rapid transmission across long-range association fibers develops significantly throughout adolescence.
14: NYT article touching on the risk of addiction with ketamine treatment and how this interacts with it often being prescribed via telemedicine. In my view, an aspect missing from this important narrative is that ketamine might have opioid properties.
15: Tiny stone points were found at a cave site in France from 54,000 years ago, suggesting that humans at that time hunted bison and horses with bows and arrows.
16: A study uses high-coverage whole-genome sequencing of 180 individuals from 12 indigenous African populations, identifying millions of novel variants and uncovering ancient population structures. Finds that the ancestors of southern African San and central African rainforest hunter-gatherers diverged from other populations over 200,000 years ago while maintaining a large effective population size.
17: A protective variant against Alzheimer's disease found in the gene MS4A4A is found to expand the population of an anti-inflammatory state of microglia.
18: A potential Alzheimer's treatment using 40 Hz light or sound to stimulate gamma rhythms in the brain has been questioned. The new study found that while the treatment slightly reduced Aβ42 production, it did not affect plaque buildup, thus failing to replicate previous results about its efficacy for reducing Alzheimer's pathology.
Part of me is now telling myself that I previously thought this finding was implausible, but I didn’t write it anywhere, so I’m probably just deluding myself to feel smarter.
19: According to a new study, well-being tends to increase even above income levels of $80,000, up to very high levels, although perhaps not the very highest ones (although those have lower sample sizes):
20: Between 2017 and 2018, there was a cluster of four cardiac surgery patients at Brigham and Women's Hospital in Boston who developed rare Mycobacterium abscessus infections. Whole-genome sequencing confirmed a link between the infections and mycobacterial growth in ice and water machines on the affected floor. It turns out that these machines had been fitted with 5.0-micron carbon filters, designed to improve taste and remove odors, that decreased the chlorine concentration from 2.5mg/L to 0mg/L in the tap water, which likely allowed the bacteria to grow.
21: The depletion of microglia or T cells prevents neurodegeneration in animal models of tauopathy. Just another piece of evidence that immunology is most likely to provide the transformative treatments that we’re looking for in neurodegenerative diseases.
22: Study finds that the inhibition of a kinase I’ve never heard of (PIKfyve) activates an unconventional protein clearance mechanism involving exocytosis of aggregation-prone proteins. This decreases ALS pathology and prolongs survival in diverse ALS cell and animal models. If it translates, it would circumvent the need to stimulate macroautophagy or the ubiquitin-proteasome system, providing a potential treatment for multiple forms of the disease.
23: Katherine Boyle asks: is suffering good? Potentially to be counterpoised with Brian Tomasik’s “Essays on Reducing Suffering”. Here’s a June 2022 update from Brian on why he doesn’t write as much as he used to.
24: I helped create a list of all the long-term care providers in cryonics and put in on the Cryonics Wiki. It contains 16 current and historical providers. I don't think there was a previous table like this.
25: New cryonics monitoring organization led by Nikki Olson aims to create dependable monitoring solutions for cryonicists, including research on sensors and devices.
26: New paper from a group at the Universidad Autónoma de Madrid
describing their method of perfusion fixation in human and non-human primate brains. They use a type of ex situ perfusion fixation, where they take the brain out of the skull, place it on the laboratory bench, and then perfuse through the two carotid arteries and the basilar artery. Importantly, they also do postfixation of 1 cm brain slabs for 24 hours after perfusion, noting that postfixation is necessary in human brains, "because perfusion-fixation is not quite optimal due to post-mortem delays and intravascular blood clots". In non-human primate brains, they also report that they can perfuse a series of sucrose solutions (5%, then 10%, then 20%) after fixative to start the cryoprotection procedure, which is a similar procedure as aldehyde-stabilized cryopreservation.
27: Retrocausality is the theoretical idea in quantum physics suggesting that future events can influence past ones. In a new manuscript, Price and Wharton propose that the statistical phenomenon of Berkson’s paradox/collider bias, when constrained by such future conditions, could be the underlying mechanism for quantum entanglement. They use this to offer a novel explanation for the seemingly causal connections observed in entangled particle pairs.
I’m not an expert in this area and I can’t evaluate the research. But because quantum physics underlies everything in our world, I think it behooves the scientifically informed observer to be aware of some of the controversies in this area, if only to recognize there are more unknown unknowns out there than we might appreciate.
#18 - I also didn't believe strobe lights were likely to help when I first heard about it, so priors confirmed I guess. I say that without having actually looked into the literature at all, so I obviously also deserve many science points /s
#19 - Reminds me of various studies on how microglia can destroy synapses in animal models of AD. In combination with the fact that women have a higher prevalence of AD than men, and women have higher rates of autoimmune issues generally, I guess I place some weight on AD eventually turning out to be some kind of microglia autoimmune disease?