<?xml version="1.0" encoding="UTF-8"?>
<rss  xmlns:atom="http://www.w3.org/2005/Atom" 
      xmlns:media="http://search.yahoo.com/mrss/" 
      xmlns:content="http://purl.org/rss/1.0/modules/content/" 
      xmlns:dc="http://purl.org/dc/elements/1.1/" 
      version="2.0">
<channel>
<title>Caroline Gachema</title>
<link>https://carolgachema.github.io/blog.html</link>
<atom:link href="https://carolgachema.github.io/blog.xml" rel="self" type="application/rss+xml"/>
<description>Biomedical Scientist &amp; Bioinformatician</description>
<generator>quarto-1.9.38</generator>
<lastBuildDate>Tue, 18 Aug 2026 21:00:00 GMT</lastBuildDate>
<item>
  <title>The Cancer With No handles</title>
  <link>https://carolgachema.github.io/blog/tnbc-deep-dive.html</link>
  <description><![CDATA[ 






<p>Breast cancer is several genuinely different diseases that happen to start in the same tissue, and the difference between them comes down to three molecular “handles” a tumor either has or doesn’t.</p>
<p>Understanding what those handles are, and what it means for a tumor to have none of them, is the only way to understand why triple-negative breast cancer is simultaneously a biology problem and, in the United States, a racial equity problem hiding in plain sight.</p>
<section id="the-positive-cancers" class="level2">
<h2 class="anchored" data-anchor-id="the-positive-cancers">The “Positive” Cancers</h2>
<p>Most breast tumors are, in a strange way, cooperative. They express receptors on their surface, molecular docking stations, that a drug can be designed to fit into like a key into a lock.</p>
<p><strong>1. Estrogen receptor positive (ER+)</strong></p>
<p>Tumors carry receptors that respond to circulating estrogen, using it as fuel for growth. This sounds like bad news, and in one sense it is, and it’s also an exploitable weakness.</p>
<p>Drugs like <em>tamoxifen</em> block the receptor directly; <em>aromatase</em> inhibitors cut off the estrogen supply at the source.</p>
<p>ER+ disease is the most common form of breast cancer, and it’s also, generally, the most survivable, precisely because the tumor has handed oncologists something to interfere with. [1][2]</p>
<p><strong>2. Progesterone receptor positive (PR+)</strong></p>
<p>Tumors work on a similar principle, often alongside ER+ status, and are treated with overlapping hormone-blocking strategies.</p>
<p><strong>3. HER2-positive</strong></p>
<p>Tumors overexpress a growth-signaling protein called HER2 on their surface, which drives aggressive and fast growth. <strong>Trastuzumab</strong> (Herceptin) and the generation of HER2-targeted drugs that followed it transformed HER2+ disease from one of the more dangerous subtypes into one of the more treatable ones, simply because there was a specific protein to aim at.</p>
<p>In every case, <em>the tumor’s molecular identity is the treatment plan.</em></p>
</section>
<section id="triple-negative" class="level2">
<h2 class="anchored" data-anchor-id="triple-negative">Triple-Negative</h2>
<p>Triple-negative breast cancer (TNBC) is defined entirely by absence. No estrogen receptor. No progesterone receptor. No HER2 amplification.</p>
<p>It accounts for roughly one in eight to one in seven breast cancer diagnoses overall, but it is not evenly distributed across age, geography, or race. [1][2]</p>
<p>TNBC skews toward younger patients, is more likely to be caught at a later stage, is more likely to recur after treatment, and its five-year survival drops sharply once the disease has spread beyond the breast. [1]</p>
<p>For a long time, the absence of a molecular target meant the absence of a targeted treatment, and chemotherapy remained the backbone of TNBC treatment for decades while other subtypes got increasingly precise weapons.</p>
<p>That’s beginning to shift:</p>
<ul>
<li><strong>Pembrolizumab</strong>, an immunotherapy drug, showed real benefit in early-stage TNBC when added to standard treatment (the KEYNOTE-522 trial). [7]</li>
<li><strong>Sacituzumab govitecan</strong>, an antibody-drug conjugate that smuggles chemotherapy directly into tumor cells via a surface protein TNBC does express, extended survival in metastatic disease (the ASCENT trial). [8]</li>
<li>And for the subset of TNBC driven by inherited BRCA1 or BRCA2 mutations, PARP inhibitors like olaparib exploit a specific DNA-repair vulnerability those mutations create (the OlympiA trial). [9]</li>
</ul>
</section>
<section id="aurora-b" class="level2">
<h2 class="anchored" data-anchor-id="aurora-b">Aurora B</h2>
<p>One of those emerging places is a small enzyme called Aurora B kinase, and understanding why requires understanding that genomic chaos is not simply bad for a tumor.</p>
<p>Every time a cell divides, its chromosomes have to be copied and pulled apart with near-perfect precision. Cancer cells are frequently sloppy at this, a phenomenon called chromosomal instability, or CIN, and the sloppiness behaves less like a symptom and more like a dial the tumor is, in effect, tuning.</p>
<p>Turn the dial too low, and the cancer stays genetically uniform: boring, predictable, an easy target for a single well-matched therapy. Turn it too high, and cells accumulate so much division-related damage that they die mid-split <em>(mitotic catastrophe)</em>.</p>
<p>The tumor’s actual advantage lives in the narrow band between those extremes, just enough chromosomal variation for natural selection to test different mutations against chemotherapy and the immune system, without so much chaos that any given cell fails to survive the division that produced it.</p>
<p>That narrow band is where drug resistance gets manufactured, where relapse originates, where metastatic clones get their start.</p>
<p>Aurora B sits directly at that dial. It’s a kinase (an enzyme that tags other proteins with phosphate groups) and depending on exactly where the tag lands, the target protein switches on, switches off, changes shape, or lets go of whatever it was holding, often a chromosome mid-division. [3][4]</p>
<p>Aurora B governs how chromosomes attach to the mitotic spindle and get pulled apart, which makes it one of the central regulators of that chaos dial itself.</p>
<p>In TNBC specifically, Aurora B runs recurrently overactive. Because the tumor is already dependent on walking a fine genomic line, pushing Aurora B activity past its tolerable limit can force TNBC cells into the mitotic catastrophe they’d otherwise avoid. [5] Combinations built around Aurora and angiogenic kinase inhibition have already reached phase II trials in previously treated, metastatic TNBC. It’s an active area of drug development. [6]</p>
<p>But a molecular vulnerability only matters for the patients who actually get diagnosed early, tested, and treated in time, and TNBC turns out to be as much a story about who gets access to that timeline as it is about kinases.</p>
</section>
<section id="why-this-lands-hardest-on-black-women" class="level2">
<h2 class="anchored" data-anchor-id="why-this-lands-hardest-on-black-women">Why This Lands Hardest on Black Women</h2>
<p>Black women in the United States are diagnosed with breast cancer at a rate roughly 5% lower than white women, according to the <em>American Cancer Society’s 2024–2025 Breast Cancer Facts and Figures report.</em> [15] Despite that, they are 38 to 41% more likely to die of breast cancer than white women — 38% by that same ACS report, 41% by ACS’s separate report on cancer among African American and Black people, and two to three times more likely to be diagnosed with TNBC specifically. [15][16][17]</p>
<p>And once diagnosed, the outcomes gap widens further. Black women with TNBC are roughly 30% more likely to die of it than white women with the same diagnosis, a gap researchers link partly to lower rates of receiving surgery and chemotherapy, meaning some of this disparity is not about the tumor at all. [12][13]</p>
<p><strong>Biologically</strong></p>
<p>Triple-negative and “basal-like” breast cancers occur at meaningfully different rates across populations with different genetic ancestry, and BRCA1 and BRCA2 mutations, which sharply raise TNBC risk, show up at higher prevalence in Black patients with metastatic TNBC than in white patients with the same diagnosis in at least one recent analysis — 12% versus 7%. [11][14]</p>
<p><strong>Structurally</strong></p>
<p>Researchers studying neighborhood deprivation have found it independently associated with TNBC risk, suggesting that where you live, and everything that correlates with that, is doing real biological work. [10]</p>
<p>Add to that well-documented gaps in access to genetic testing, to timely screening, to surgery, and to chemotherapy once diagnosed, and the picture that emerges is one where biology sets part of the stage, and a healthcare system that delivers unequal access does a great deal of the rest.</p>
<p>Neither thread fully explains the gap alone. Both are true at once, which makes this a harder problem to solve.</p>
</section>
<section id="the-molecular-work-still-matters" class="level2">
<h2 class="anchored" data-anchor-id="the-molecular-work-still-matters">The Molecular Work Still Matters</h2>
<p>None of the structural half of this problem gets fixed by a kinase. But the biological half is a starting point.</p>
<hr>
<section id="notes" class="level3">
<h3 class="anchored" data-anchor-id="notes">Notes</h3>
<ol type="1">
<li>Howard, F. et al.&nbsp;“Epidemiology of Triple-Negative Breast Cancer: A Review” — Cancer Journal. TNBC incidence and survival gap vs.&nbsp;hormone-receptor-positive disease.</li>
<li>American Cancer Society, “Triple-negative Breast Cancer” — receptor status, staging, and general prognosis.</li>
<li>Krenn, V. &amp; Musacchio, A. “The Aurora B Kinase in Chromosome Bi-Orientation and Spindle Checkpoint Signaling” — Frontiers in Cell and Developmental Biology.</li>
<li>Zaytsev, A. et al.&nbsp;“Aurora B kinase is recruited to multiple discrete kinetochore and centromere regions” — Journal of Cell Biology.</li>
<li>Yang, J. et al.&nbsp;“Antineoplastic effects of an Aurora B kinase inhibitor in breast cancer” — Molecular Cancer. Barasertib (AZD1152) mechanism and preclinical breast cancer data.</li>
<li>Diamond, J. et al.&nbsp;“A phase II clinical trial of the Aurora and angiogenic kinase inhibitor ENMD-2076 for previously treated, advanced, or metastatic triple-negative breast cancer” — Breast Cancer Research.</li>
<li>Schmid, P. et al.&nbsp;“Pembrolizumab for Early Triple-Negative Breast Cancer” (KEYNOTE-522) — New England Journal of Medicine, 2020.</li>
<li>Bardia, A. et al.&nbsp;“Sacituzumab Govitecan in Metastatic Triple-Negative Breast Cancer” (ASCENT) — New England Journal of Medicine, 2021.</li>
<li>Tutt, A.N.J. et al.&nbsp;“Adjuvant Olaparib for Patients with BRCA1- or BRCA2-Mutated Breast Cancer” (OlympiA) — New England Journal of Medicine, 2021.</li>
<li>Understanding the role of neighborhood deprivation in racial disparities in triple negative breast cancer — Cancer, 2025.</li>
<li>Sisters Network Inc.&nbsp;“About Breast Cancer” — subtype distribution by race, HR+/HER2- prevalence by group.</li>
<li>Matar-Ujvary, R. et al.&nbsp;“Triple-Negative Breast Cancer, Recurrence, and Survival Outcomes in Young Black Women” — Journal of Racial and Ethnic Health Disparities, 2025.</li>
<li>“Triple-negative breast cancer in Black women: Risks and outlook” — Medical News Today, 2025.</li>
<li>Biomarker, treatment patterns, and survival differences in metastatic triple-negative breast cancer by race in the United States — npj Breast Cancer, 2026.</li>
<li>Black Cancer Research Foundation (BCRF). “Black Women and Breast Cancer: Disparities and Research.”</li>
<li>“Triple-Negative Breast Cancer in Black Women” — Black Health, citing American Cancer Society Cancer Facts &amp; Figures for African American/Black People.</li>
<li>Racial and regional disparities of triple negative breast cancer incidence rates in the United States — an analysis of SEER/NPCR data, 2011–2019.</li>
</ol>


</section>
</section>

 ]]></description>
  <category>oncology</category>
  <category>health equity</category>
  <category>cancer biology</category>
  <guid>https://carolgachema.github.io/blog/tnbc-deep-dive.html</guid>
  <pubDate>Tue, 18 Aug 2026 21:00:00 GMT</pubDate>
  <media:content url="https://carolgachema.github.io/blog/images/aurorab.jpg" medium="image" type="image/jpeg"/>
</item>
<item>
  <title>Magic Shrooms</title>
  <link>https://carolgachema.github.io/blog/magics-shrooms.html</link>
  <description><![CDATA[ 






<p>A Mazatec healer named <strong>María Sabina</strong> was administering mushrooms the old way, at night, by candlelight and in a ceremony she called a <em>velada</em>, chanting over people who came to her sick.</p>
<p>In 1955, a New York banker and amateur mycologist named <strong>R. Gordon Wasson</strong> talked his way into one of her ceremonies and became, <em>by his own account</em>, the first outsider to eat the mushrooms and write down what happened. His account ran in <em>Life</em> two years later under a very interesting title “Seeking the Magic Mushroom” and the name stuck to the plant for good. It cost Sabina dearly, obviously. The flood of seekers her sudden fame set off got her run out of her own village.</p>
<p>In a supervised, and even unsupervised session today, a person lies down, puts on an eyeshade, and listens to a curated playlist while they go on their shroom journey. What they describe afterwards…</p>
<p>The usual boundary between self and everything else goes soft, there’s a sense of contact with something larger than themselves, sometimes a visit from someone they’ve lost. Run that protocol on people with a life-threatening diagnosis specifically, and the numbers get hard to look away from, a single high dose, and roughly seven in ten participants later rank the night among the five most meaningful experiences of their entire life, up there with the birth of a child. In particular, the fear of dying tends not to come back the way it was.</p>
<p>None of it, as far as anyone can tell, was built for us. The genes a mushroom uses to make <strong>psilocybin</strong> turn up scattered across unrelated species of dung and wood-rotting fungi, the signature of genes that jumped between species. Also the leading explanation for why any fungus bothered is almost insulting. The compound looks like an insect deterrent, evolved to make the mushroom taste bad, or think badly, to the beetles that would otherwise eat it before it spread its spores.</p>
<p>Once swallowed, <strong>psilocybin</strong> converts within minutes to <strong>psilocin</strong>, which doesn’t stop at the receptor responsible for the visions. It walks directly into TrkB and wedges itself into a channel running through the receptor’s own membrane-spanning segment, prying the whole structure into a shape roughly a thousand times more sensitive to its natural signal than basic antidepressants medicinal chemists spent decades engineering to do exactly that job.</p>
<p><strong>What the drug does next</strong></p>
<p>Blocks that one pocket, and the antidepressant-like effects disappear, while the hallucinations carry on completely unbothered.</p>
<div class="quarto-figure quarto-figure-center">
<figure class="figure">
<p><img src="https://carolgachema.github.io/blog/images/mexicana_blog.jpg" class="img-fluid figure-img"></p>
<figcaption><em>Psilocybe mexicana</em> fruiting bodies in habitat. Photo via Wikimedia Commons/ CC BY 3.0.</figcaption>
</figure>
</div>
<p><strong>TrkA and TrkB</strong> are cousins by gene duplication, products of the same ancient tyrosine-kinase lineage, and they still show it. Line up their kinase domains and roughly six residues in every ten are identical. Their AlphaFold-predicted structures and the two practically trace each other outline for outline, especially through the transmembrane segment.</p>
<p>This is the exact region where a 2023 study out of <strong>Eero Castrén’s lab</strong> in Helsinki pinpointed psilocin’s binding pocket on TrkB. TrkA answers to a different growth factor, nerve growth factor rather than BDNF, but once activated it fires the identical downstream relay. Which is the same <em>MAPK cascade</em>, the same <em>PI3K/Akt survival signaling</em>, the same <em>calcium-mobilizing PLCγ pathway</em> that TrkB uses to build new synapses.</p>
<p>Structurally and functionally, TrkA is close enough to TrkB that biology has trouble telling them apart in some contexts.</p>
<p><strong>Pancreatic ductal adenocarcinoma</strong> doesn’t behave like most solid tumors. Instead of waiting for a blood vessel or a lymphatic channel to hitch a ride out of the pancreas, it grows into nerves directly, a pattern called <em>perineural invasion.</em> This shows up in the overwhelming majority of resected specimens. Cancer cells chemotax toward the nearest nerve fascicle, breach its protective sheath, and travel inside it as a corridor</p>
<div class="quarto-figure quarto-figure-center">
<figure class="figure">
<p><img src="https://carolgachema.github.io/blog/images/perineural_invasion.png" class="img-fluid figure-img"></p>
<figcaption>Histological slide (H&amp;E stain) demonstrating perineural invasion (PNI), with malignant tumor glands encircling a nerve bundle (N). Image via NIH Open-i</figcaption>
</figure>
</div>
<p>It’s also why pancreatic cancer produces some of the most severe pain in solid-tumor oncology.</p>
<p>And the whole process runs on the same molecular grammar as <em>neuroplasticity</em> — the tumor secretes nerve growth factor, activates TrkA on both itself and the adjacent nerve, and the nerve responds by sprouting toward the tumor in return. Each side is, in the most literal biochemical sense, growing toward the other.</p>
<p>When researchers sequenced the whole genomes of early-stage tumors in 2012, mutations in the genes that guide axons toward their targets turned out to be among the most recurrent alterations in the cancer’s genome, as fundamental to its identity as the <em>KRAS</em> mutations that drive nearly every case.</p>
<p>Trkb isn’t confined to the nerves running through the tumor’s edges. It’s expressed on the pancreatic cancer cells themselves, where it helps cancer cells let go of their epithelial identity and resist <strong>anoikis</strong>, the death a cell is normally sentenced to the moment it loses its grip on surrounding tissue.</p>
<p><strong>Anoikis resistance</strong> is the property a cancer cell needs to survive the trip through the bloodstream and found a metastasis somewhere else. So the receptor that psilocin was shown to sensitize with is expressed on the tumor cells, directly relevant to how the cancer spreads.</p>
<p>The results described earlier (the sustained relief, the fear of dying loosening its grip) held up again in a second, independent trial published the same week in 2016, this time out of NYU. Patients with pancreatic cancer are, if anything, the population with the most to gain from that kind of intervention, carrying some of the heaviest psychological burden of any cancer diagnosis. Taking that option away from them on the basis of an untested structural argument would be its own kind of harm.</p>
<p><strong>Over the Last decade</strong></p>
<ul>
<li><p>Cancer neuroscience has been building the case that nerves are not passive bystanders in tumor growth but active participants recruited by the tumor itself.</p></li>
<li><p>Psychiatric neuroscience has discovered that its most promising new therapy works by directly rewiring a growth-factor receptor.</p></li>
</ul>
<p>Whether concurrent TrkA and TrkB activation in a pancreatic tumor produces nothing worth worrying about is mostly unknown. amplifies the invasion the tumor already relies on, or actually gets in its own way, is, well, mostly unknown. All three outcomes are mechanistically defensible, and we can only hope for…</p>
<div class="quarto-figure quarto-figure-center">
<figure class="figure">
<p><img src="https://carolgachema.github.io/blog/images/tumor_shrooms.png" class="img-fluid figure-img"></p>
<figcaption>Schematic comparing tumor innervation (left) and perineural invasion (right), illustrating nerve recruitment versus cellular invasion of the perineurium. Image created with BioRende</figcaption>
</figure>
</div>
<hr>
<p><strong>Further Reading</strong></p>
<p>Wasson, R.G. (1957, May 13). Seeking the magic mushroom. <em>Life</em>, 42(19), 100–120.</p>
<p>Reynolds, H.T., Vijayakumar, V., Gluck-Thaler, E., et al.&nbsp;(2018). Horizontal gene cluster transfer increased hallucinogenic mushroom diversity. <em>Evolution Letters</em>, 2(2), 88–101. https://doi.org/10.1002/evl3.42</p>
<p>Moliner, R., Girych, M., Brunello, C.A., et al.&nbsp;(2023). Psychedelics promote plasticity by directly binding to BDNF receptor TrkB. <em>Nature Neuroscience</em>, 26(6), 1032–1041. https://doi.org/10.1038/s41593-023-01316-5</p>
<p>Bapat, A.A., Hostetter, G., Von Hoff, D.D., &amp; Han, H. (2011). Perineural invasion and associated pain in pancreatic cancer. <em>Nature Reviews Cancer</em>, 11, 695–707. https://doi.org/10.1038/nrc3131</p>
<p>Biankin, A.V., Waddell, N., Kassahn, K.S., et al.&nbsp;(2012). Pancreatic cancer genomes reveal aberrations in axon guidance pathway genes. <em>Nature</em>, 491(7424), 399–405. https://doi.org/10.1038/nature11547</p>
<p>Griffiths, R.R., Johnson, M.W., Carducci, M.A., et al.&nbsp;(2016). Psilocybin produces substantial and sustained decreases in depression and anxiety in patients with life-threatening cancer. <em>Journal of Psychopharmacology</em>, 30(12), 1181–1197. https://doi.org/10.1177/0269881116675513</p>
<p>Ross, S., Bossis, A., Guss, J., et al.&nbsp;(2016). Rapid and sustained symptom reduction following psilocybin treatment for anxiety and depression in patients with life-threatening cancer. <em>Journal of Psychopharmacology</em>, 30(12), 1165–1180. https://doi.org/10.1177/0269881116675512</p>
<p>Wiesmann, C., Ultsch, M.H., Bass, S.H., &amp; de Vos, A.M. (1999). Crystal structure of nerve growth factor in complex with the ligand-binding domain of the TrkA receptor. <em>Nature</em>, 401(6749), 184–188. https://doi.org/10.1038/43705</p>
<p><strong>Notes</strong></p>
<p><em>This post is adapted from a longer manuscript, “Convergent Neurotrophic Signaling,” proposing a preclinical framework for investigating this interface directly.</em></p>



 ]]></description>
  <category>neuroscience</category>
  <category>oncology</category>
  <category>psychedelics</category>
  <category>pharmacology</category>
  <guid>https://carolgachema.github.io/blog/magics-shrooms.html</guid>
  <pubDate>Fri, 31 Jul 2026 21:00:00 GMT</pubDate>
  <media:content url="https://carolgachema.github.io/blog/images/magic_shrooms.jpg" medium="image" type="image/jpeg"/>
</item>
<item>
  <title>On Congenital Insensitivity</title>
  <link>https://carolgachema.github.io/blog/on-congenital-insensitivity.html</link>
  <description><![CDATA[ 






<p>When most people think about losing the ability to feel pain, they imagine an invulnerability story. Pop culture treats congenital insensitivity to pain as a flawed superpower, a biological quirk that lets you step on glass or hold hot coals without flinching. Better yet, a heart that never breaks.</p>
<p><strong>Alexander Chesler</strong> and <strong>Carsten Bönnemann</strong> began examining a handful of patients with rare, loss of function mutations in a gene called <em>PIEZO2</em>. When these patients were tested, their pain thresholds came back completely normal, a pinprick still felt sharp, a hot stove still hurt. What they’d actually lost wasn’t pain at all. It was the subconscious software that tells the brain where the body lives in space.</p>
<div class="quarto-figure quarto-figure-center">
<figure class="figure">
<p><img src="https://carolgachema.github.io/blog/images/muscle-spindle-3d.jpg" class="img-fluid figure-img"></p>
<figcaption>Schematic of muscle spindle mechanoreceptors and sensory nerve pathways. Image via Wikimedia Commons / OpenStax, CC BY 4.0.</figcaption>
</figure>
</div>
<p><strong>Split sensation into <em>PIEZO2</em> constituent physics.</strong></p>
<p>We are used to thinking of <em>touch</em> as a single sense, but the nervous system treats temperature, chemical irritants, and mechanical force as entirely distinct languages.</p>
<p>We’ve known the molecular receptors responsible for heat and cold work that earned <strong>David Julius</strong> part of the 2021 Nobel Prize in Physiology or Medicine. But the hardware responsible for sensing physical force remained elusive until <strong>Ardem Patapoutian,</strong> sharing that same Nobel Prize, identified the PIEZO family of ion channels.</p>
<p><em>PIEZO2</em> is a massive, three-bladed, propeller-shaped protein embedded in the membranes of our sensory neurons.</p>
<p>When your skin is stretched, poked, or compressed, or when a muscle spindle lengthens as you bend your elbow, the cell membrane physically pulls open the blades of the <em>PIEZO2</em> channel.</p>
<p>Positively charged ions rush into the cell, firing an electrical signal to the brain. Force is converted into thought in microseconds.</p>
<div class="column-body-outset">
<div style="width: 100%; margin: 2.5em 0;">

  <div id="piezo2-viewer-placeholder" style="width: 100%; height: 550px; border-radius: 10px; background: linear-gradient(135deg, #0E2A42, #1c3a52); display: flex; flex-direction: column; align-items: center; justify-content: center; cursor: pointer;" onclick="loadPiezo2Viewer()">
    <span style="color: #fff; font-family: Atkinson Hyperlegible, sans-serif; font-weight: 700; font-size: 1rem; letter-spacing: 0.04em; border: 1.5px solid #fff; border-radius: 999px; padding: 0.8rem 1.6rem;">▶ Load interactive 3D structure</span>
  </div>

  <div id="piezo2-viewer-frame" style="display: none; width: 100%; height: 550px; border-radius: 10px; overflow: hidden;"></div>

  <p style="font-size: 0.85em; color: #666; text-align: center; margin-top: 0.8em;">
    <strong>Interactive 3D structure:</strong> mouse PIEZO2 mechanosensitive channel, cryo-EM (PDB 6KG7; Wang et al., 2019, <em>Nature</em>). Click and drag to rotate the propeller blades; scroll to zoom.
  </p>

</div>

<script>
function loadPiezo2Viewer() {
  document.getElementById('piezo2-viewer-placeholder').style.display = 'none';
  var frame = document.getElementById('piezo2-viewer-frame');
  frame.style.display = 'block';
  var iframe = document.createElement('iframe');
  iframe.src = 'https://molstar.org/viewer/?pdb=6KG7&hide-controls=1';
  iframe.style.width = '100%';
  iframe.style.height = '100%';
  iframe.style.border = 'none';
  iframe.title = '3D interactive structure of PIEZO2';
  frame.appendChild(iframe);
}
</script>
</div>
<p>In humans born without functional <em>PIEZO2</em>, the consequences reveal just how much invisible work mechanical sensation does behind the scenes.</p>
<p>If you ask a person with a <em>PIEZO2</em> mutation to close their eyes and touch their nose with an index finger, they cannot do it.</p>
<p>Without visual confirmation, their brain loses track of where their hands are. If you blindfold them and gently move their foot up or down, they cannot tell you which direction it moved, or if it moved at all. They lack proprioception (the “sixth sense” generated by continuous mechanical feedback from muscle spindles and joint receptors).</p>
<p>To walk, they must look directly at their feet, manually substituting visual data for the automated subconscious telemetry their brain ought to be receiving from their legs. If you turn off the lights in a room, they will sway and fall, stripped of their internal coordinates in the dark.</p>
<p>Intriguingly, their pain sensation is almost eerily ordinary. They feel the burn of a hot stove and the sting of chili pepper at normal thresholds, because thermal and chemical nociceptors run on separate molecular machinery (like the TRPV family).</p>
<p>But tested against a <em>pinprick</em> or a <em>blunt pressure probe</em>, their mechanical pain thresholds come back indistinguishable from anyone else’s. This surprised even the researchers, since it meant the everyday sense of <em>“ouch”</em> doesn’t route through <em>PIEZO2</em> at all. Somewhere in the skin, a separate circuit of pain fibers is doing that job on its own.</p>
<p>However, <em>PIEZO2</em> doesn’t control whether touch turns into pain after an injury.</p>
<p>Healthy nervous systems develop <em>tactile allodynia</em> after a burn or a bruise. Patients without <em>PIEZO2</em> never develop it. Light touch stays harmless, even on inflamed tissue, because the channel that would normally convert that touch into pain signaling simply isn’t there.</p>
<p>Chronic pain remains one of the most stubborn failures of modern pharmacology, largely because targeting broad pain pathways often numbs the entire nervous system or carries high addiction liabilities. <em>PIEZO2</em> selectively encodes mechanical force and opens up hyper-targeted therapeutic possibilities, drugs that selectively dampen mechanical hypersensitivity in conditions like <em>fibromyalgia</em> or <em>neuropathic pain</em> without wiping out a patient’s ability to feel thermal danger or basic touch.</p>
<p><em>PIEZO2</em> is actively guiding how our bodies assemble themselves. Patients with <em>PIEZO2</em> deficiency frequently develop <em>severe scoliosis, joint hypermobility</em>, and <em>skeletal deformities</em>. During development, the spine and joints rely on mechanical feedback from movement to align correctly. Without the constant tension readings supplied by <em>PIEZO2</em>, the body grows without knowing its own geometry.</p>
<p>Which brings us back to what sensory perception actually is.</p>
<p>We tend to view our senses as external tools, cameras and microphones taking in an outside world. But it’s also about defining the internal self. <strong>Literally</strong> It is the molecular architecture of presence, constantly anchoring your mind inside a physical structure.</p>
<p>Stripped of that single protein, the brain is simply left floating in space.</p>
<hr>
<p><strong>Further Reading</strong></p>
<p>Chesler, A.T., Szczot, M., Bharucha-Goebel, D., et al.&nbsp;(2016). The Role of PIEZO2 in Human Mechanosensation. <em>The New England Journal of Medicine</em>, 375(14), 1355–1364. https://doi.org/10.1056/NEJMoa1602812</p>
<p>Nagi, S.S., Marshall, A.G., Makdani, A., et al.&nbsp;(2019). An ultrafast system for signaling mechanical pain in human skin. <em>Science Advances</em>, 5(7), eaaw1297. https://doi.org/10.1126/sciadv.aaw1297</p>
<p>Szczot, M., Liljencrantz, J., Ghitani, N., et al.&nbsp;(2018). PIEZO2 mediates injury-induced tactile pain in mice and humans. <em>Science Translational Medicine</em>, 10(462), eaat9892. https://doi.org/10.1126/scitranslmed.aat9892</p>
<p>Coste, B., Mathur, J., Schmidt, M., et al.&nbsp;(2010). Piezo1 and Piezo2 are essential components of distinct mechanically-activated cation channels. <em>Science</em>, 330(6000), 55–60. https://doi.org/10.1126/science.1193270</p>
<p>Woo, S.H., Lukacs, V., de Nooij, J.C., et al.&nbsp;(2015). PIEZO2 is the principal mechanotransducer of proprioception in mice. <em>Nature Neuroscience</em>, 18(12), 1756–1762. https://doi.org/10.1038/nn.4162</p>
<p><strong>Notes</strong></p>
<p><em>Banner photo: Cultured dorsal root ganglion (DRG) sensory neuron explant showing axonal outgrowth. Image via <a href="https://commons.wikimedia.org/wiki/File:DRG_Chicken_e7.jpg">Wikimedia Commons</a>, CC BY-SA 3.0.</em></p>



 ]]></description>
  <category>neuroscience</category>
  <category>mechanobiology</category>
  <category>sensory physiology</category>
  <guid>https://carolgachema.github.io/blog/on-congenital-insensitivity.html</guid>
  <pubDate>Thu, 30 Jul 2026 21:00:00 GMT</pubDate>
  <media:content url="https://carolgachema.github.io/blog/images/DRG_Chicken_e7.jpg" medium="image" type="image/jpeg"/>
</item>
<item>
  <title>Ping Pong, Ping Pong</title>
  <link>https://carolgachema.github.io/blog/ping-pong-ping-pong.html</link>
  <description><![CDATA[ 






<p>In 2022, a team based in Melbourne connected a layer of living neurons to a video game and watched them learn to play it. <strong>Actual neurons</strong>, roughly 800,000 of them, growing on a dish, with no eyes, no body, and no reward system anyone had given them. At least not on purpose.</p>
<p>The setup is called <em>DishBrain,</em> built by a team at Cortical Labs led by neuroscientist <em>Brett Kagan.</em> The neurons were a mix of mouse cortical cells and human neurons derived from stem cells. They were cultured directly onto a high-density array of electrodes, a grid dense enough to both listen to the electrical activity of individual neurons and deliver precise electrical stimulation to them cell by cell.</p>
<p>The team wired this array into a simulated version of <strong>Pong</strong>. The position of the ball was converted into a stimulation pattern delivered to one region of the dish. The neurons’ own resulting activity in another region was read out and converted into paddle movement.</p>
<p>This is much stranger than a standard reinforcement learning setup because, what were the neurons were working with when they got something wrong?</p>
<p>A game-playing AI is normally trained by maximizing reward, like points for winning or a number going up. Neurons in a dish don’t have access to anything like a dopamine signal you can just hand them. So the team, working with theoretical neuroscientist <em>Karl Friston,</em> a co-author on the paper, built the feedback around <em>Friston’s</em> free energy principle instead, <strong>the idea that biological systems act, fundamentally, to minimize surprise about their environment rather than to chase reward directly.</strong></p>
<ul>
<li><p>When the dish hit the ball, it received predictable, structured stimulation.</p></li>
<li><p>When it missed, it received unstructured electrical noise.</p></li>
</ul>
<p>Instead of earning a prize laying well earned a more orderly, a less chaotic world to be stimulated by.</p>
<p>Within about five minutes of gameplay, the culture’s rallies got longer. The neurons weren’t just twitching randomly or occasionally connecting with the ball by chance, their hit rate <em>improved</em> over the course of play, in a way the researchers could track and reproduce.</p>
<p>Interestingly, the <em>human-derived neurons</em> sustained <em>longer</em> rallies than the <em>mouse-derived</em> ones, though the team has been careful not to lean too hard on that difference given how early this work still is.</p>
<p>Kagan himself <em>ALSO</em> stressed that DishBrain shows no evidence of consciousness, and the original paper’s use of the word <em>“sentience”</em> in its title, meaning simply the capacity to sense and respond to an environment, drew real pushback from other neuroscientists who felt it invited exactly the kind of overclaiming Kagan says he wanted to avoid.</p>
<p>A system can behave adaptively, in a measurable, repeatable way, without that behavior implying anything about what it’s like, if anything, to be that system.</p>
<p>Deep reinforcement learning agents typically need enormous numbers of training episodes to master something as simple as Pong. A follow-up study directly comparing biological neuron cultures with deep RL agents on the same task found the neurons could reach comparable performance using far less experience, a property researchers call sample efficiency.</p>
<p>Whatever the neurons are doing to organize themselves around this task, they’re doing it with less data and drastically less power than a GPU cluster needs.</p>
<p><strong>Real translational pull:</strong> a living, adaptable neural system on a dish is a plausible platform for testing how drugs, or conditions like epilepsy, alter learned behavior directly, without ever needing an animal model in the loop.</p>
<p>What this suggests, I think, is the capacity to act so as to make your surroundings more predictable isn’t something that waits around for a full brain, a body, or a nervous system shaped by millions of years of evolution to show up. Whatever learning fundamentally is, it may be running on much simpler hardware than we’ve been assuming.</p>
<hr>
<p><strong>Further reading</strong></p>
<p>Kagan, B.J., Kitchen, A.C., Tran, N.T., et al.&nbsp;(2022). In vitro neurons learn and exhibit sentience when embodied in a simulated game-world. <em>Neuron</em>. https://doi.org/10.1016/j.neuron.2022.09.001</p>
<p>Friston, K. (2010). The free-energy principle: a unified brain theory? <em>Nature Reviews Neuroscience</em>, 11, 127–138.</p>
<p>Biological Neurons Compete with Deep Reinforcement Learning in Sample Efficiency in a Simulated Gameworld (2024). arXiv:2405.16946.</p>



 ]]></description>
  <category>computational neuroscience</category>
  <category>neurotechnology</category>
  <category>learning and intelligence</category>
  <guid>https://carolgachema.github.io/blog/ping-pong-ping-pong.html</guid>
  <pubDate>Tue, 14 Jul 2026 21:00:00 GMT</pubDate>
  <media:content url="https://carolgachema.github.io/blog/images/dishbrain-electrode-array.jpg" medium="image" type="image/jpeg"/>
</item>
<item>
  <title>Built for the Deep</title>
  <link>https://carolgachema.github.io/blog/built-for-the-deep.html</link>
  <description><![CDATA[ 






<p>For over a thousand years, the <em>Bajau</em> people of <em>Southeast Asia</em> have lived mostly at sea, moving between the waters of Indonesia, Malaysia, and the Philippines in houseboats, gathering food by freediving with nothing but a set of weights and wooden goggles.</p>
<p>Some spend up to five hours a day underwater.</p>
<p>Divers routinely descend past 60 metres and hold their breath for several minutes at a time, in a range most people could never approach even with training.</p>
<div class="quarto-figure quarto-figure-center">
<figure class="figure">
<p><img src="https://carolgachema.github.io/blog/images/bajau-houses-1923.jpg" class="img-fluid figure-img"></p>
<figcaption>Sama-Bajau houses in Cawa Cawa, Zamboanga City, 1923. Public domain (National Council of the Protestant Episcopal Church Department of Missions), via Wikimedia Commons.</figcaption>
</figure>
</div>
<p>That has always been treated as a story about skill and lifestyle. Then, in 2018, a team led by geneticist <strong>Melissa Ilardo</strong> set out to test whether any of it was written into the Bajau’s biology itself, rather than just trained into it.</p>
<p>The team compared the Bajau to a nearby population, <strong>the Saluan</strong>, who share a similar environment but don’t dive for a living. The finding was striking. Bajau spleens were significantly larger than the Saluan’s, and the difference held even among Bajau who don’t dive. That detail matters because it meant this wasn’t a body adapting to a lifestyle through training. It was something the Bajau were already carrying, <em>genetically</em>, before they ever entered the water.</p>
<p>The <em>spleen</em> turns out to be the key organ here.</p>
<p>In diving mammals, it contracts during a dive, releasing a reserve of oxygen-rich red blood cells into circulation, buying extra time underwater. The researchers traced the Bajau’s enlarged spleens to variation in a gene called <em>PDE10A</em>, which appears to influence thyroid hormone levels known to regulate spleen size.</p>
<p>They also found evidence of selection on a second gene, <em>BDKRB2</em>, linked to the body’s diving reflex, the automatic narrowing of blood vessels in the limbs that redirects oxygen toward vital organs during breath holding. Put together, the Bajau seem to carry a genuine, inherited physiological toolkit for life underwater.</p>
<p>Both of those genes point toward the same organ, even though, well, neither one sits inside it. The dive reflex exists to protect the two structures that can least afford to run out of oxygen, the heart and the brain.</p>
<p>When <em>BDKRB2</em> tightens blood vessels in the arms and legs, it’s rerouting blood toward the brain specifically, the organ most vulnerable to <em>hypoxia</em> and the one whose damage is hardest to reverse. The reflex itself is coordinated by circuits in the brainstem, which means the apparatus doing the redirecting sits inside the very organ it evolved to protect.</p>
<p>It’s one of the clearest documented cases we have of a population evolving a specific, measurable adaptation to a specific way of life, and it has real medical relevance.</p>
<p><strong>Beyond Curiosity</strong></p>
<p>Understanding how the body naturally tolerates low oxygen has direct implications for treating acute <em>hypoxia</em> in emergency and critical care medicine, including the neuroprotective strategies used when the brain itself is starved of oxygen after a stroke or cardiac arrest, situations where every minute of unprotected hypoxia costs measurable brain tissue.</p>
<p>A population that has spent a thousand years surviving repeated, voluntary bouts of exactly that stress is, in effect, a natural experiment in the same question critical care medicine is still trying to solve artificially.</p>
<p><strong>Many stories like this never get found</strong></p>
<p>The Bajau’s variant was only discoverable because someone chose to sequence their genomes in the first place. Most of the world hasn’t had that chance.</p>
<p>The <strong>GWAS Catalog</strong>, one of the largest public repositories of genomic association data, is still roughly 95 percent European ancestry, even though people of European ancestry make up less than 9 percent of the world’s population.</p>
<p>It means population-specific variants, protective or otherwise, in African, Asian, Latin American, and Indigenous populations are largely invisible to the reference data modern medicine is built on. Simply because the sequencing was never done.</p>
<p>Most of the world still waiting for a desicion to be made about them. Inclusion</p>
<hr>
<p><strong>Further reading</strong></p>
<p>Ilardo, M.A., Moltke, I., Korneliussen, T.S., et al.&nbsp;(2018). Physiological and Genetic Adaptations to Diving in Sea Nomads. <em>Cell</em>, 173(3), 569–580. https://doi.org/10.1016/j.cell.2018.03.054</p>
<p>Petrovski, S. &amp; Goldstein, D.B. (2016). Unequal representation of genetic variation across ancestry groups creates healthcare inequality in the application of precision medicine. <em>Genome Biology</em>, 17, 157.</p>
<p><strong>Notes</strong></p>
<p><em>Banner photo: A Sama-Bajau village in Pulau Omadal, Sabah, Malaysia, by Tom Gunnar Hoogervorst (2012), CC BY 3.0, via Wikimedia Commons.</em></p>



 ]]></description>
  <category>population genetics</category>
  <category>genomic equity</category>
  <category>evolutionary biology</category>
  <guid>https://carolgachema.github.io/blog/built-for-the-deep.html</guid>
  <pubDate>Tue, 30 Jun 2026 21:00:00 GMT</pubDate>
  <media:content url="https://carolgachema.github.io/blog/images/bajau-village-omadal.jpg" medium="image" type="image/jpeg"/>
</item>
</channel>
</rss>
