<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>Map the Brain blog</title>
    <link>https://connectome.quest/blog/</link>
    <atom:link href="https://connectome.quest/blog/feed.xml" rel="self" type="application/rss+xml" />
    <description>What is new in Pyr and the other brain mapping projects of the Seung Lab, Princeton University.</description>
    <language>en</language>
    <lastBuildDate>Sat, 21 Nov 2026 19:44:41 GMT</lastBuildDate>
    <item>
      <title>Lost Fragments 2.0: The Glia Strike Back</title>
      <link>https://connectome.quest/blog/lost-fragments-2-the-glia-strike-back/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/lost-fragments-2-the-glia-strike-back/</guid>
      <pubDate>Sat, 21 Nov 2026 19:44:41 GMT</pubDate>
      <dc:creator>Amy R. Sterling</dc:creator>
      <description>The lab's full list of 11,605 untraced retina fragments is in Pyr, played from the center of the retina outward. Plus a new way to tag glia.</description>
      <content:encoded><![CDATA[<p>Lost Fragments just got a lot bigger.</p>
<h2 id="what-are-lost-fragments">What are Lost Fragments?<a class="blog-anchor" href="#what-are-lost-fragments" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>The retina we are mapping in Pyr is full of pieces nobody is tracking yet. Some are whole cells that were never registered, and the rest are large pieces of neurite cut off from their cells. Tidy one up and complete it, and it counts as a completed cell.</p>
<h2 id="the-lab-s-full-list">The lab's full list<a class="blog-anchor" href="#the-lab-s-full-list" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>When Lost Fragments opened on November 20, it had a first sample of 151 fragments. Now the lab has sent us their full list: <strong>11,605 fragments</strong>. They left out anything their filters took for glia, anything edited a lot recently, and every cell already in the proofreading sheets, so you are never redoing someone's work.</p>
<p>They come in three groups:</p>
<ul><li><strong>Easy</strong> (2,735): each probably has a cell body, which makes the cell easy to recognize and complete.</li><li><strong>Medium</strong> (5,273): large pieces of neurite with no cell body.</li><li><strong>Hard</strong> (3,597): the smallest pieces. Fewer clues, so they can be harder to place.</li></ul>
<p>Play hands them out from the center of the retina outward, and passes over any fragment someone else has already started. Some fragments come with a clue: if a piece was cut off a cell in the lab's sheets, Pyr tells you which cell, so you can check where it belongs.</p>
<h2 id="the-glia-strike-back">The glia strike back<a class="blog-anchor" href="#the-glia-strike-back" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>The lab's glia filter is still early, so some glia slipped through. If your fragment is a glia, press <strong>It's a glia</strong> instead of proofreading it. Each glia is one column rising up, covered in little barb-like projections. Say whether it is one glia or several fused together, put the crosshairs in a trunk, then submit. That point tells the lab exactly where the glia is.</p>
<p>No proofreading needed, and it counts as a Quest. We'll stash it away for the Glia Quest, launching in early December, where you will split fused glia apart.</p>
<h2 id="also-new">Also new<a class="blog-anchor" href="#also-new" aria-label="Link to this section" title="Link to this section">#</a></h2>
<ul><li><strong>Your work saves as you go.</strong> Each fragment you hold keeps its checklist, your annotations and your view, even after a reload or on another computer.</li><li><strong>Later</strong> keeps a fragment in My Cells and gives you a new one. You can hold up to 5 at a time.</li><li><strong>Nicknames.</strong> Give any cell your own name, like &quot;cell I merged a branch into&quot;. Only you see it.</li></ul>
<h2 id="how-to-play">How to play<a class="blog-anchor" href="#how-to-play" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Open Pyr, go to Retina, press Play and pick Lost Fragments. You need edit access on Retina: <a href="https://connectome.quest/blog/how-to-unlock-edit-access-in-pyr/" target="_blank" rel="noopener noreferrer">here is how to unlock it</a>.</p>
<p>Thank you for mapping with us. For Science!</p>
<p>Amy</p>]]></content:encoded>
    </item>
    <item>
      <title>How to make Pyr run faster</title>
      <link>https://connectome.quest/blog/how-to-make-pyr-run-faster/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/how-to-make-pyr-run-faster/</guid>
      <pubDate>Sat, 21 Nov 2026 01:47:10 GMT</pubDate>
      <dc:creator>Amy R. Sterling</dc:creator>
      <description>Simple settings to keep Pyr smooth, a look at how Pyr loads a cell, and what we have already sped up.</description>
      <content:encoded><![CDATA[<p>Pyr feeling slow? A setting or two usually fixes it.</p>
<p>Every cell you proofread helps map the brain, one neuron at a time. The images behind each dataset are far too large for any computer to hold at once, so Pyr streams in just the pieces you are looking at, while the lab's servers keep track of every edit every player makes. That makes Pyr a challenge not only to play but also to deliver, and we are always working to make it faster.</p>
<p>Pyr draws a lot at once: sharp 2D images, detailed 3D cells, chat, and a few festive extras. A fast computer handles that easily, but an older laptop or a slow connection can start to drag. Below are the settings that help most, biggest wins first. After them comes a quick look at how Pyr loads a cell, which helps you tell <em>what</em> is slow, and a rundown of what we've already done to speed things up.</p>
<p>To open Settings, press <strong>Shift+S</strong>. You can type any setting's name into the search box at the top to jump straight to it.</p>
<h2 id="1-lower-the-2d-image-detail">1. Lower the 2D image detail<a class="blog-anchor" href="#1-lower-the-2d-image-detail" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>If the 2D images take a long time to sharpen, start here.</p>
<p>Go to <strong>Settings &gt; 2D images &gt; 2D image detail</strong> and pick:</p>
<ul><li><strong>Half (faster loading):</strong> loads about a quarter of the data. For most tracing, this is still plenty sharp.</li><li><strong>Quarter (fastest, blurry):</strong> loads about a sixteenth of the data. Good for a very slow connection.</li></ul>
<p>The images stay at that detail until you choose <strong>Full</strong> again, so switch back whenever you need a close look at a tricky spot.</p>
<h3 id="a-different-detail-for-each-dataset">A different detail for each dataset<a class="blog-anchor" href="#a-different-detail-for-each-dataset" aria-label="Link to this section" title="Link to this section">#</a></h3>
<p>Some datasets are heavier than others, so Pyr can remember a detail for each one. Right below <strong>2D image detail</strong> there is a line named after the dataset you are on, such as <strong>On MEC</strong>. Leave it at <strong>Same as above</strong>, or pick <strong>Full</strong>, <strong>Half</strong> or <strong>Quarter</strong> just for that dataset.</p>
<p>Pyr saves the choice to your account and switches the 2D detail by itself when you change datasets. For example, you might use <strong>Half</strong> on MEC for heavy proofreading and keep <strong>Full</strong> on Cilia.</p>
<h2 id="2-load-2d-images-from-the-closer-server">2. Load 2D images from the closer server<a class="blog-anchor" href="#2-load-2d-images-from-the-closer-server" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Retina, MEC and the Sandbox have a second copy of their 2D images hosted in Asia. Other datasets always load from the Americas.</p>
<p>In <strong>Settings &gt; 2D images &gt; 2D images load from</strong>, <strong>Automatic</strong> times both copies once and uses the faster one for you. The menu shows which one it picked. If you know the other copy is closer to you, choose <strong>Asia</strong> or <strong>Americas</strong> yourself.</p>
<figure class="blog-post__figure"><img src="https://connectome.quest/assets/images/blog/pyr-speed/settings-2d-images-v2.webp" alt="Settings, 2D images: 2D image detail and 2D images load from" loading="lazy" /><figcaption>Settings &gt; 2D images</figcaption></figure>
<h2 id="3-keep-fewer-cells-in-the-3d-view">3. Keep fewer cells in the 3D view<a class="blog-anchor" href="#3-keep-fewer-cells-in-the-3d-view" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Every cell shown in 3D uses graphics memory, and a view full of finished cells is one of the most common reasons Pyr slows down.</p>
<ul><li>Take cells out of the view once you are done with them.</li><li><strong>Ctrl+Shift+X</strong> clears every cell from the view at once. Your cells and your edits stay exactly as they are; they are just no longer drawn.</li></ul>
<h2 id="4-quiet-the-screen">4. Quiet the screen<a class="blog-anchor" href="#4-quiet-the-screen" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>A few settings in <strong>Settings &gt; Screen</strong> cut down on extra drawing:</p>
<ul><li><strong>Calm screen: no animations or glows</strong> stops the animations and glows around text, buttons and panels. The cells and the 3D view stay as they are. A few small animations follow after a reload.</li><li><strong>Particles</strong> off: your points go down without the little burst of sparkles.</li><li>During Thanksgiving week, <strong>Show the Thanksgiving turkey</strong> turns off the turkey that wanders along the bottom of the screen. The turkey button by the chat box does the same, and Calm screen keeps him away too.</li></ul>
<figure class="blog-post__figure"><img src="https://connectome.quest/assets/images/blog/pyr-speed/settings-screen-v2.webp" alt="Settings, Screen: Particles and Calm screen" loading="lazy" /><figcaption>Settings &gt; Screen</figcaption></figure>
<p>In <strong>Settings &gt; 3D looks</strong>, turning off <strong>Turn cells slowly</strong> stops the 3D view from slowly spinning by itself.</p>
<figure class="blog-post__figure"><img src="https://connectome.quest/assets/images/blog/pyr-speed/settings-3d-looks-v2.webp" alt="Settings, 3D looks: Turn cells slowly" loading="lazy" /><figcaption>Settings &gt; 3D looks</figcaption></figure>
<h2 id="5-tune-the-viewer-settings">5. Tune the Viewer Settings<a class="blog-anchor" href="#5-tune-the-viewer-settings" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>For finer control, go to <strong>Settings &gt; Neuroglancer</strong> and click <strong>Viewer Settings</strong>. Pyr remembers what you set here on your account, so it stays the same on every new cell.</p>
<figure class="blog-post__figure"><img src="https://connectome.quest/assets/images/blog/pyr-speed/settings-neuroglancer-v2.webp" alt="Settings, Neuroglancer: the Viewer Settings button" loading="lazy" /><figcaption>Settings &gt; Neuroglancer &gt; Viewer Settings</figcaption></figure>
<ul><li><strong>Enable prefetching:</strong> loads nearby images before you reach them. Turning it off saves data on a slow connection, though scrolling to a new spot may take a moment longer.</li><li><strong>Concurrent chunk requests:</strong> how many pieces of data load at the same time. A lower number can help on a weak connection.</li><li><strong>GPU memory limit</strong> and <strong>System memory limit:</strong> how much memory the viewer may use. If Pyr stutters or the browser tab crashes, try a lower number. If you have a strong computer and cells keep fading in and out as you move, a higher number can help.</li></ul>
<figure class="blog-post__figure"><img src="https://connectome.quest/assets/images/blog/pyr-speed/viewer-settings-panel-v2.webp" alt="Viewer Settings: GPU memory limit, System memory limit, Concurrent chunk requests and Enable prefetching" loading="lazy" /><figcaption>The Viewer Settings panel</figcaption></figure>
<h2 id="6-check-whether-it-s-you-or-the-servers">6. Check whether it's you or the servers<a class="blog-anchor" href="#6-check-whether-it-s-you-or-the-servers" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Sometimes the slowdown isn't your computer at all. Open your Profile and go to <strong>Datasets</strong>. The <strong>Server status</strong> line at the top says whether the lab servers for Retina, the Sandbox and MEC are OK, Slow or Down, and it opens by itself when one is having trouble. If a server is slow, it's not your setup, and things usually pick up again soon.</p>
<figure class="blog-post__figure"><img src="https://connectome.quest/assets/images/blog/pyr-speed/server-status-v2.webp" alt="Profile, Datasets: the Server status line" loading="lazy" /><figcaption>Profile &gt; Datasets &gt; Server status</figcaption></figure>
<h2 id="how-pyr-works-behind-the-scenes">How Pyr works behind the scenes<a class="blog-anchor" href="#how-pyr-works-behind-the-scenes" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>When you open a cell, Pyr is talking to several servers at once. Knowing who does what makes it easier to tell what is slow.</p>
<p><strong>The 2D images.</strong> A dataset's microscope images are far too big to download whole, so they are stored as millions of small blocks called chunks, each a little cube of image. Every dataset is saved at several levels of detail, from sharp to very blurry. Pyr works out which chunks cover what's on your screen, shows a blurry version first so you see something right away, then fills in the sharp chunks. As you scroll through slices, it also fetches the images just ahead of you. This is the part that depends most on your internet connection, and it's what the 2D image detail and image source settings change.</p>
<p><strong>The 3D cells.</strong> Each cell is built from many small pieces. When you select a cell, Pyr first asks the lab's proofreading server which pieces belong to it, then downloads the 3D shape of each piece and draws them together. A big cell can be made of thousands of pieces, which is why it fills in bit by bit. Once they arrive, every shape stays in your graphics card's memory while it is in view, which is why clearing finished cells helps so much.</p>
<p><strong>Your edits.</strong> A merge or a cut goes to the proofreading server, which updates the cell and builds its new 3D shape. That's the &quot;Updating the 3D shape&quot; you sometimes see after an edit.</p>
<p><strong>Cell information.</strong> For the cells in view, Pyr also asks the lab's servers whether each one has been proofread and what type of cell it is.</p>
<p><strong>The game itself.</strong> Chat, claims, the Cell Library, leaderboards, Achievements and notifications live on Pyr's own game server, separate from the science data.</p>
<p>So different kinds of slow have different causes:</p>
<ul><li><strong>2D images stay blurry:</strong> usually your connection, or the image server. Try a lower image detail or the closer server.</li><li><strong>3D cells take a long time to appear, or an edit takes a while:</strong> usually the proofreading server. Check Server status.</li><li><strong>Panning and turning feel choppy:</strong> usually too much for the graphics to draw at once. Clear finished cells, turn on Calm screen, or lower the memory limits.</li></ul>
<h2 id="what-we-ve-already-sped-up">What we've already sped up<a class="blog-anchor" href="#what-we-ve-already-sped-up" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>A lot of the work on Pyr this fall has gone into making it lighter. Some highlights:</p>
<ul><li><strong>Faster 2D images.</strong> Pyr now asks for fewer image chunks, in a better order: fewer blurry backup levels, only a short way fetched ahead instead of dozens of blocks, and a piece that fails to arrive is tried again. In our tests, Sandbox images that took about 8 seconds to fill in now take about 5.</li><li><strong>Copies of the images in Asia</strong> for Retina, MEC and the Sandbox, with Automatic picking the faster one for you.</li><li><strong>Fewer questions to the lab's servers.</strong> Cell status is now asked for many cells in one request instead of one at a time, and remembered for a short while. After an edit, Pyr only looks up the claims on your screen.</li><li><strong>Smoother big views.</strong> Views with hundreds of cells or thousands of annotation points draw and respond faster, and cut points appear the moment you place them, even on very big cells.</li><li><strong>A faster Cell Library.</strong> It loads cells in pages instead of all at once and does much less work when it updates. The Mine list no longer freezes when you have finished many cells, and the Cilia Quest list loads much faster.</li><li><strong>Lighter chat and leaderboards.</strong> When many people play at once, joining and leaving now reach chat as one small message each. The leaderboard refreshes less often and asks for far less, and the bell asks only for what is new.</li><li><strong>Your Viewer Settings stick.</strong> Memory and loading choices are saved on your account and kept for every new cell.</li><li><strong>Server status</strong> in your Profile, so you can see when the slowness is on our side.</li></ul>
<h2 id="still-slow">Still slow?<a class="blog-anchor" href="#still-slow" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Tell us in chat or send us a bug report, and mention which dataset you were on and what felt slow. Every report helps us make Pyr lighter for everyone.</p>
<p>Happy proofreading!</p>]]></content:encoded>
    </item>
    <item>
      <title>Cilia Status Guide</title>
      <link>https://connectome.quest/blog/cilia-status-guide/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/cilia-status-guide/</guid>
      <pubDate>Thu, 05 Nov 2026 19:19:08 GMT</pubDate>
      <dc:creator>Celia D</dc:creator>
      <description>Hello Cilia Hunters! This is a quick guide to each cilia status and when to use it. Click each one to see an example. Happy Hunting! @[gallery] Cilium Annotated - Cilium emerges from the cell, both centrioles and cilium centerline are annotated. Cilium Annotated (fixation artifact) - Cilium…</description>
      <content:encoded><![CDATA[<p>Hello Cilia Hunters!</p>
<p>This is a quick guide to each cilia status and when to use it. Click each one to see an example. Happy Hunting!</p>
<div class="blog-post__gallery"><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/5e3d8942-ca4b-43d7-b813-d6de53f312d8.png" alt="Cilium Annotatedhttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6134465776582656 - Cilium emerges from the cell, both centrioles and cilium centerline are annotated." loading="lazy" /><figcaption><strong><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6134465776582656" target="_blank" rel="noopener noreferrer">Cilium Annotated</a></strong> - Cilium emerges from the cell, both centrioles and cilium centerline are annotated.</figcaption></figure><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/1a72811d-b6be-45ee-ae62-4643cb493797.png" alt="Cilium Annotated fixation artifacthttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6224937215852544 - Cilium structure is visible near the surface of the cell, but appears attached/internal to the soma. This is caused by membranes fusing during data collection." loading="lazy" /><figcaption><strong><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6224937215852544" target="_blank" rel="noopener noreferrer">Cilium Annotated (fixation artifact)</a></strong> - Cilium structure is visible near the surface of the cell, but appears attached/internal to the soma. This is caused by membranes fusing during data collection.</figcaption></figure><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/716741d0-03ec-4427-ac24-99b1930ae4e2.webp" alt="Need Help: One Centriole Missinghttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6697415730003968 - You found one centriole, but the other is MIA. Cilium may or may not be present." loading="lazy" /><figcaption><strong><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6697415730003968" target="_blank" rel="noopener noreferrer">Need Help: One Centriole Missing</a></strong> - You found one centriole, but the other is MIA. Cilium may or may not be present.</figcaption></figure><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/5f85693d-c38a-479f-ad23-835913f5c015.png" alt="No Cilium BB &amp; DC clearhttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5742473875816448 - No cilium, but centrioles are visible and clearly differentiated basal body closer to the surface." loading="lazy" /><figcaption><strong><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5742473875816448" target="_blank" rel="noopener noreferrer">No Cilium (BB &amp; DC clear)</a></strong> - No cilium, but centrioles are visible and clearly differentiated (basal body closer to the surface).</figcaption></figure><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/42e20ebf-19c9-43ea-9155-c4af9690b696.png" alt="No cilium BB &amp; DC unclearhttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5185996505219072 - No cilium, centriole positions make it unclear which is the basal body and which is the daughter." loading="lazy" /><figcaption><strong><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5185996505219072" target="_blank" rel="noopener noreferrer">No cilium (BB &amp; DC unclear)</a></strong> - No cilium, centriole positions make it unclear which is the basal body and which is the daughter.</figcaption></figure><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/f5256b80-d077-4da6-8560-7a9f5494af2a.png" alt="Skipped low qualityhttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6280972446203904 -- in this case the soma is missing." loading="lazy" /><figcaption><strong><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6280972446203904" target="_blank" rel="noopener noreferrer">Skipped (low quality)</a></strong> -- in this case the soma is missing.</figcaption></figure></div>
<p>Other statuses:</p>
<ul><li><strong>Need Help</strong> - You can't find any cilium or centrioles, or there's another issue that stops you from annotating the cell. Good to leave clarifying notes in most cases!</li><li><strong>No Cilium</strong> - Nothing found. This should only be used after review (use Need Help if you are the first annotator).</li></ul>
<h2 id="tip-distinguishing-bb-and-dc">Tip: Distinguishing BB and DC<a class="blog-anchor" href="#tip-distinguishing-bb-and-dc" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>You found two centrioles, but how do you know which one is the basal body, and which is the daughter?</p>
<p><strong>Answer</strong>: The basal body is closer to the soma's surface. In cases where there is a cilium, it is always the centriole at the base of the cilium. When no cilium is present, it will still often be close (or closer than the daughter) to the soma membrane.</p>
<p>Here are two images where we can tell a difference between the centrioles:</p>
<div class="blog-post__pair"><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/956d07e1-4e76-40fa-96c4-e012b6161811.png" alt="In this examplehttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5116782905917440, the basal body yellow annotation is clearly closer to the surface. The centrioles can appear as different shapes depending on their orientation." loading="lazy" /><figcaption><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5116782905917440" target="_blank" rel="noopener noreferrer">In this example</a>, the basal body (yellow annotation) is clearly closer to the surface. The centrioles can appear as different shapes depending on their orientation.</figcaption></figure><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/488a3842-59e6-46c3-9863-be1d4a849e36.png" alt="Here is another examplehttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5663507412418560 where we can clearly see which is the basal body yellow annotation." loading="lazy" /><figcaption><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5663507412418560" target="_blank" rel="noopener noreferrer">Here is another example</a> where we can clearly see which is the basal body (yellow annotation).</figcaption></figure></div>
<p><em>I think one may be closer but I'm not sure. What do I do?</em></p>
<p>If the centrioles are not easily distinguishable, consider them &quot;No cilium (BB and DC unclear).&quot;</p>
<p>Here are two examples where it is not easy to differentiate the centrioles:</p>
<div class="blog-post__pair"><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/af995631-ad71-46b9-80fb-e40cdc2a4d5d.png" alt="In this examplehttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5160372340260864, although both centrioles were found, it is unclear which is which. You should still use one of each annotation type to mark them - red and yellow." loading="lazy" /><figcaption><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5160372340260864" target="_blank" rel="noopener noreferrer">In this example</a>, although both centrioles were found, it is unclear which is which. You should still use one of each annotation type to mark them - red and yellow.</figcaption></figure><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/c5ca69f9-13ec-4372-9aab-531f735ae793.png" alt="Here is a second examplehttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5136791346610176 where the centrioles are not clearly distinguished." loading="lazy" /><figcaption><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6242682812760064" target="_blank" rel="noopener noreferrer">Here is a second example</a> where the centrioles are not clearly distinguished.</figcaption></figure></div>]]></content:encoded>
    </item>
    <item>
      <title>How Do We See? The Mysteries Hiding in Plain Sight</title>
      <link>https://connectome.quest/blog/how-do-we-see/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/how-do-we-see/</guid>
      <pubDate>Thu, 05 Nov 2026 14:37:07 GMT</pubDate>
      <dc:creator>Amy R. Sterling</dc:creator>
      <description>Some of the biggest mysteries of vision begin in a thin layer of cells at the back of your eye. Here is the known, the unknown and how you can help map the brain to make neuroscience history.</description>
      <content:encoded><![CDATA[<p>Every time you open your eyes, billions of photons enter your pupils. A fraction of a second later, you're seeing the world in color, depth, and motion. You can recognize a face across a crowded room, follow a bird in flight, or notice something moving at the edge of your vision.</p>
<p>How does your brain accomplish all this?</p>
<p>Surprisingly, scientists are still figuring it out. And some of the biggest mysteries begin in a thin layer of cells at the back of your eye.</p>
<h2 id="your-retina-is-a-little-piece-of-brain">Your retina is a little piece of brain<a class="blog-anchor" href="#your-retina-is-a-little-piece-of-brain" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Light enters your eye through the cornea and pupil. The lens focuses it onto the retina, a delicate sheet of neural tissue lining the back of the eye.</p>
<p>Here, specialized cells called rods and cones detect light and convert it into electrical signals. Those signals pass through an intricate network of neurons before traveling along the optic nerve to the brain.</p>
<p>But the retina does much more than detect light. It helps identify edges, changes in brightness, and movement. Some retinal neurons even respond preferentially when objects move in a particular direction.</p>
<p>Your eyes are already making sense of the world before your brain gets the message.</p>
<figure class="blog-post__figure"><picture><source media="(max-width: 700px)" srcset="https://connectome.quest/assets/images/blog/bipolar-cells/eye-diagram-phone.webp" /><img src="https://connectome.quest/assets/images/blog/bipolar-cells/eye-diagram.webp" alt="How we see: the eye, with a close look at the retina's cells" loading="lazy" /></picture></figure>
<h2 id="a-forest-of-neurons">A forest of neurons<a class="blog-anchor" href="#a-forest-of-neurons" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>The retina contains many different kinds of neurons, arranged in layers. Bipolar cells relay information from photoreceptors, amacrine cells shape and modify those signals, and ganglion cells carry the results to the brain.</p>
<p>Their shapes are extraordinary. Some stretch vertically between layers, while others extend elaborate branches sideways, gathering information from their neighbors.</p>
<figure class="blog-post__figure"><picture><source media="(max-width: 700px)" srcset="https://connectome.quest/assets/images/blog/bipolar-cells/real-retina-cells-phone.webp" /><img src="https://connectome.quest/assets/images/blog/bipolar-cells/real-retina-cells.webp" alt="How a signal flows through the retina: photoreceptors, a bipolar cell, a starburst amacrine cell and a ganglion neuron, real cells from Eyewire II at their true depths" loading="lazy" /></picture></figure>
<p>The arrangement of these branches matters. Which cells connect, where they connect, and how strongly they communicate all help determine what information reaches the brain.</p>
<h2 id="what-don-t-we-know">What don't we know?<a class="blog-anchor" href="#what-don-t-we-know" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Quite a lot, actually.</p>
<p>Scientists are still working to identify all the different types of retinal neurons and understand what each contributes to vision. We know some of the circuitry responsible for detecting motion, contrast, and changes in light, but we don't yet understand how all those circuits work together when we look at a complex, constantly changing world.</p>
<p>And there's an especially tantalizing question: If we could map every neuron and every connection in the retina, could we predict what the circuit sees?</p>
<p>A wiring diagram alone isn't enough. Neurons are living, dynamic cells. But combining their anatomy with recordings of their activity will bring us to many new discoveries. That possibility is what makes modern connectomics so exciting.</p>
<h2 id="a-million-tiny-clues">A million tiny clues<a class="blog-anchor" href="#a-million-tiny-clues" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>With advances in electron microscopy and artificial intelligence, scientists can now reconstruct neurons in extraordinary detail.</p>
<p>Pyr is one such prjoject. It includes several connectomics datasets, including Eyewire II, a volume of mouse retina containing ~100,000cells. For hundreds of its neurons, researchers have both detailed anatomy and recordings of how the cells respond to visual stimuli.</p>
<p>That gives us a rare opportunity to connect the shape of a neuron with what it actually does. And there are countless connections left to investigate.</p>
<h2 id="help-us-discover-what-s-hiding-in-the-retina">Help us discover what's hiding in the retina<a class="blog-anchor" href="#help-us-discover-what-s-hiding-in-the-retina" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Mapping neural circuits is painstaking work. AI can trace neurons through microscopic images, but it makes mistakes. Human eyes are remarkably good at spotting where a branch has gone astray.</p>
<p>That's where you come in.</p>
<p>In Pyr, the new citizen science game from the creators of Eyewire, players help map real neurons from neuroscience datasets, including the EyeWire II retina.</p>
<p>Every correction contributes to a more accurate map. And every better map gives scientists new opportunities to investigate how neural circuits work.</p>
<p>We can't promise what discoveries lie ahead. Perhaps we'll uncover clues about how we see or make progress toward solving issues of the eye such as blindness. One thing is for certain: this is uncharted territory, and mapping it is only possible with your help!</p>
<p>That's the lovely thing about exploring the unknown: you don't get to know what you'll find before you begin.</p>
<p>Come explore the brain with us. Play Pyr and help unravel the mysteries of vision!</p>
<p class="blog-post__cta"><a class="blog-post__button" href="https://pyr.ai/" target="_blank" rel="noopener noreferrer">Play Pyr</a></p>]]></content:encoded>
    </item>
    <item>
      <title>Guide to Mapping Bipolar Cells (BCs)</title>
      <link>https://connectome.quest/blog/guide-to-mapping-bipolar-cells-bcs/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/guide-to-mapping-bipolar-cells-bcs/</guid>
      <pubDate>Thu, 05 Nov 2026 03:14:15 GMT</pubDate>
      <dc:creator>Amy R. Sterling</dc:creator>
      <description>Every image you see passes through a bipolar cell. Learn what these relay cells do and how to proofread one in Pyr.</description>
      <content:encoded><![CDATA[<p>Every image you have ever seen passed through a bipolar cell. These small and tidy cells sit in the middle of the retina. In Eyewire II/ Pyr's retina dataset, they are some of the cells you will encounter most frequently. Here is what they do and how to proofread one.</p>
<h2 id="what-bipolar-cells-do">What bipolar cells do<a class="blog-anchor" href="#what-bipolar-cells-do" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Light hits the photoreceptors (the rods and cones) at the back of the retina. The photoreceptors can't talk to the brain directly. They hand their signal to bipolar cells, and the bipolar cells pass it on to both amacrine and ganglion cells. Ganglion axons form the optic nerve and send a pre-processed signal carrying information about objects and movement from the eye to the brain.</p>
<figure class="blog-post__figure"><picture><source media="(max-width: 700px)" srcset="https://connectome.quest/assets/images/blog/bipolar-cells/eye-diagram-phone.webp" /><img src="https://connectome.quest/assets/images/blog/bipolar-cells/eye-diagram.webp" alt="How we see: the eye, with a close look at the retina's cells" loading="lazy" /></picture></figure>
<p>Bipolar cells are a critical relay step, and the direct path from photoreceptor to neuron.</p>
<figure class="blog-post__figure"><picture><source media="(max-width: 700px)" srcset="https://connectome.quest/assets/images/blog/bipolar-cells/real-retina-cells-phone.webp" /><img src="https://connectome.quest/assets/images/blog/bipolar-cells/real-retina-cells.webp" alt="How a signal flows through the retina: photoreceptors, a bipolar cell, a starburst amacrine cell and a ganglion neuron, real cells from Eyewire II at their true depths" loading="lazy" /></picture></figure>
<div class="blog-post__video"><iframe src="https://www.youtube-nocookie.com/embed/u7TOLm2HXRk" title="YouTube video" loading="lazy" allow="accelerometer; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen referrerpolicy="strict-origin-when-cross-origin"></iframe></div>
<p>Bipolar cells are also part of EyeWire history. EyeWire players reconstructed bipolar cells and the starburst amacrine cells they connect to. The work showed that bipolar types with different timing connect to different parts of the starburst cell, a wiring pattern that helps the retina tell which way something is moving, which became the first ever Eyewire publication in the prestigious journal Nature back in 2014 (Kim et al., Nature).</p>
<p>Today, Eyewire II players are carrying that work forward, and you have already mapped more than 10,000 cells together:</p>
<div class="blog-post__video"><iframe src="https://www.youtube-nocookie.com/embed/LUYDh1yROzg" title="YouTube video" loading="lazy" allow="accelerometer; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen referrerpolicy="strict-origin-when-cross-origin"></iframe></div>
<p>Every bipolar cell you proofread in Pyr adds to that map. If you'd like to learn more about the science, scroll to the bottom of the post. Below find instructions on how to proofread Bipolar Cells (BCs).</p>
<h2 id="claiming-a-bc">Claiming a BC<a class="blog-anchor" href="#claiming-a-bc" aria-label="Link to this section" title="Link to this section">#</a></h2>
<ol><li>Head to the <strong>Retina Dataset</strong> (if you don't have access yet, see <a href="https://connectome.quest/blog/how-to-unlock-edit-access-in-pyr/" target="_blank" rel="noopener noreferrer">this post</a>).</li><li>Open the <strong>Cell Library</strong>, navigate to <strong>Available Cells</strong>, and <strong>Claim</strong> one.</li><li><strong>Proofread</strong> the cell (instructions below)</li><li>When you have fully proofread the BC, mark it <strong>Complete</strong></li></ol>
<h2 id="how-to-proofread-a-bc">How to proofread a BC<a class="blog-anchor" href="#how-to-proofread-a-bc" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Check the whole cell, branch by branch:</p>
<ul><li><strong>Dendrites:</strong> follow each one up to its tip. Look for pieces that broke off (they need a merge) and pieces of other cells stuck on (they need a cut).</li><li><strong>Axon:</strong> follow it down to the terminals, and check that the terminals all belong to this cell.</li><li><strong>Each branch end:</strong> a bipolar cell has many small ends. Check every one.</li></ul>
<figure class="blog-post__figure"><img src="https://connectome.quest/assets/images/blog/bipolar-cells/real-bipolar-cell.webp" alt="A real bipolar cell proofread in Eyewire II, with its dendrites, cell body, axon and axon terminals labeled" loading="lazy" /></figure>
<h2 id="keep-track-of-what-you-checked">Keep track of what you checked<a class="blog-anchor" href="#keep-track-of-what-you-checked" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>A bipolar cell has lots of small branch ends, and for proofreading accuracy you must mark each one as you finish it. <strong>You may use Highlight OR Annotations</strong>. Or both if you're crazy. Use whichever tool you like best. The only annotation that is required is &quot;Can't Complete&quot;, which should be used to mark any spots on the branches that cannot be traced due to issues like EM image quality.</p>
<p><strong>Highlight Mode</strong></p>
<ul><li>Turn on Highlight Mode from its icon in the top bar.</li><li><strong>Ctrl+click</strong> two points on the cell, and the stretch between them is marked with a highlighter stroke.</li><li><strong>To soma</strong> marks the way from a point back to the cell body, and the button beside it marks every branch beyond a point.</li><li>Pick a color, or add your own, and press <strong>Undo</strong> if you mark the wrong piece.</li></ul>
<p><strong>Annotations</strong></p>
<ul><li>Make an annotation layer and pick the point tool.</li><li><strong>Ctrl+click</strong> to drop a point on each branch end you have checked.</li><li>Change the layer's color and point size so your points are easy to see.</li><li>Points are listed in the layer's panel: <strong>right click</strong> one to jump to it, or step through them with the arrows.</li></ul>
<p>Some people like Highlight because it colors the branch itself. Others like annotation points because they can step through them one by one later. Either way, when every end is marked, you know the cell is done and you can mark it complete and claim another. If you get stuck, you can ask for help in Chat or via the Help feature in the Cell Library.</p>
<p>Happy proofreading, and thank you for helping map the brain!</p>
<h2 id="more-science">More Science!<a class="blog-anchor" href="#more-science" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Check out the cast of the retina <a href="https://amyleesterling.github.io/retina/cast/" target="_blank" rel="noopener noreferrer">here</a>.</p>
<p>Each bipolar cell has two ends, which is where the name comes from:</p>
<ul><li><strong>Dendrites</strong> reach up to the photoreceptors and collect their signal.</li><li><strong>An axon</strong> runs down into the inner retina and ends in a spray of <strong>axon terminals</strong>, where it passes the signal on.</li></ul>
<figure class="blog-post__figure"><picture><source media="(max-width: 700px)" srcset="https://connectome.quest/assets/images/blog/bipolar-cells/bipolar-cell-diagram-phone.webp" /><img src="https://connectome.quest/assets/images/blog/bipolar-cells/bipolar-cell-diagram.webp" alt="A bipolar cell, drawn and labeled, beside a real one from Eyewire II" loading="lazy" /></picture></figure>
<h3 id="why-they-matter">Why they matter<a class="blog-anchor" href="#why-they-matter" aria-label="Link to this section" title="Link to this section">#</a></h3>
<p>Bipolar cells don't just pass the picture along. They split it up.</p>
<ul><li><strong>ON and OFF.</strong> Some bipolar cells respond when light gets brighter (ON), others when it gets darker (OFF). In the mouse, types 1 to 4 are OFF, and types 5 to 9 and the rod bipolar cell are ON.</li><li><strong>About 15 types.</strong> Each type reports a different piece of the scene: fast or slow changes, light or dark, fine detail or broad strokes. The rod bipolar cell carries vision in dim light.</li><li><strong>Layers.</strong> Each type ends its axon at its own depth in the inner retina, so the signals stay sorted into separate channels, like wires in a cable.</li></ul>
<p>These parallel channels are the raw material for everything the retina computes before the brain sees anything. So if you think about it, your BCs are helping to map BCs and uncovering how mammalian eyes see. Pretty cool!</p>]]></content:encoded>
    </item>
    <item>
      <title>Mapping MEC</title>
      <link>https://connectome.quest/blog/mapping-mec/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/mapping-mec/</guid>
      <pubDate>Wed, 28 Oct 2026 22:56:54 GMT</pubDate>
      <dc:creator>Amy R. Sterling</dc:creator>
      <description>The lab needs help proofreading axons!</description>
      <content:encoded><![CDATA[<h3 id="the-lab-needs-help-proofreading-axons">The lab needs help proofreading axons!<a class="blog-anchor" href="#the-lab-needs-help-proofreading-axons" aria-label="Link to this section" title="Link to this section">#</a></h3>
<p>Here is how:</p>
<ol><li><strong>Claim a cell</strong></li></ol>
<p>Open the Cell Library on MEC, click Available, pick a cell and press Claim. It is yours until you complete or release it.</p>
<ol><li><strong>Find the AIS</strong></li></ol>
<p>The <em>AIS (axon initial segment)</em> is where the axon leaves the cell body. It is the starting line for everything else.</p>
<p>Under your cell in the Cell Library, open Marks and pick AIS. Press Go to the lab’s AIS point. The lab has already placed a point where they believe the axon starts. Check it. The axon usually leaves the soma (or a thick dendrite close to it) as a single smooth, thin branch with no spines. If the lab’s point is on it, good. If it is off, move the crosshairs onto the start of the axon. Press Confirm AIS here. The AIS mark goes where the crosshairs are.</p>
<p>If the cell has no distinct axon, skip the mark and say so in your notes when you complete it.</p>
<ol><li><strong>Proofread the axon (</strong>and only the axon)</li></ol>
<p>Follow the axon from the AIS out to every one of its ends. Along the way, fix what is wrong.</p>
<ul><li>Missing pieces: the branch stops in 3D but clearly goes on in the image. Find the next piece and merge it on (press M, then Ctrl+click both pieces).</li><li>Something that does not belong: a piece of another cell joined on. Cut it off (press C, Ctrl+click red points on one side of the join and blue points on the other, then press Submit cut).</li></ul>
<p>You may cut the axon off from the rest of the cell during proofreading. Just make sure to merge is back before you complete the cell.</p>
<p>Leave the rest of the cell alone. There is no need to fix or check dendrites, spines and the cell body. We're investigating the downstream partners of these cells, which are reached through synapses on the axons, so that's why we need these first.</p>
<p><strong>Do not merge mitochondria</strong>. Inside the axon you will see small dark ovals, often as separate pieces &lt;image&gt;. They are mitochondria, the powerhouse of the cell. We plan to analyze them later, so please keep their meshes separate.</p>
<p>Some axons run out past the edge of the imaged volume. Proofread up to the edge and move on.</p>
<p>Stuck on a branch? If a branch should go on but you cannot find the next piece, ask for help in the Cell Library or in chat. You can share your view and carry on in other areas until you hear back.</p>
<p>Keeping track. On a big axon it is easy to lose your place. <strong>Highlight</strong> (Shift+H) lets you colour the stretches you have already checked. You can also use Lantern mode to hide sections of the cell so that you can focus on specific branches.</p>
<p>The other annotation layer in Marks (True End, Can’t Fix and Hits Edge) are optional now. Use them if they help you keep track, but only the AIS is required.</p>
<ol><li><strong>Complete it</strong></li></ol>
<p>When every end of the axon is done, Complete the cell, either in the Cell Library or with the Δ button next to its segment ID, and choose Axon complete. It counts as one of your completed cells, and the lab’s spreadsheet records it. Proofread axons count toward you cell completion, even if eventually we go back and add more to the dendritic arbor.</p>
<h3 id="merge-errors">Merge Errors<a class="blog-anchor" href="#merge-errors" aria-label="Link to this section" title="Link to this section">#</a></h3>
<p>MEC is currently experiencing some errors with splits and merges. This is a dataset error rather than an issue with the game interface.</p>
<p>For now please follow this protocol: If you encounter errors please add new annotation layers named &quot;Merge Errors&quot; and &quot;Split Errors&quot; and use the line annotation type to mark where you want the split or merge. You can then continue to complete the rest of the cell as normal. Please leave a note that you encountered errors and what type (merge, split, both) when you submit the cell as complete.</p>
<p><em>Need help?</em></p>
<p>Use chat, ask in the Help tab, or drop us an email.</p>
<p><em>Need a bigger challenge</em>? Try <a href="https://connectome.quest/blog/post.html?p=super-quests-mec-inhibitory-interneurons" target="_blank" rel="noopener noreferrer">Super Quests</a> :)</p>
<p>Thank you for mapping neurons in Pyr. Every axon you finish goes straight into the connectome, bringing us one branch closer to understanding the brain. For science!</p>]]></content:encoded>
    </item>
    <item>
      <title>Super Quests: MEC Inhibitory Interneurons</title>
      <link>https://connectome.quest/blog/super-quests-mec-inhibitory-interneurons/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/super-quests-mec-inhibitory-interneurons/</guid>
      <pubDate>Wed, 28 Oct 2026 17:59:36 GMT</pubDate>
      <dc:creator>Amy R. Sterling</dc:creator>
      <description>Some neurons ask for an afternoon. These could keep you exploring for weeks.

We’re introducing Super Quests in Pyr: a chance to help map enormous inhibitory interneurons in the medial entorhinal cortex (MEC), a brain region involved in navigation and memory.

These cells have sprawling branches…</description>
      <content:encoded><![CDATA[<p>Some neurons <a href="https://connectome.quest/blog/post.html?p=mapping-mec" target="_blank" rel="noopener noreferrer">ask for an afternoon</a>. These could keep you exploring for weeks.</p>
<p>We’re introducing Super Quests in Pyr: a chance to help map enormous inhibitory interneurons in the medial entorhinal cortex (MEC), a brain region involved in navigation and memory.</p>
<p>The medial entorhinal cortex helps the brain keep track of where you are. It is home to grid cells, which fire in a regular pattern as an animal moves through space. Interneurons are the cells that hold that activity in check: they inhibit their neighbors, and the giant ones reach a great many of them.</p>
<p>The new Super Quest features <strong>giant</strong> interneurons of the MEC. Each has so many branches that it can take weeks to finish. Nobody has mapped these cell types before. You can be among the first to see one whole and help us decipher what they do!</p>
<p><strong>A precise question: who do they inhibit?</strong></p>
<p>The lab has selected 29 cells and grouped them by their predicted targets:</p>
<ul><li>Pyramidal Target — 17 cells: predicted to inhibit pyramidal neurons exclusively.</li><li>Non-Pyramidal Target — 12 cells: predicted to inhibit non-pyramidal cells exclusively.</li></ul>
<p>Those are predictions. Your proofreading will help us reconstruct these neurons and analyze their connections to find out whether the predictions hold.</p>
<h3 id="how-a-super-quest-works">How a Super Quest works<a class="blog-anchor" href="#how-a-super-quest-works" aria-label="Link to this section" title="Link to this section">#</a></h3>
<p>Open the Cell Library on MEC and pick the <strong>Super Quests</strong> chip. You may claim one at a time and keep it for up to two months (if you need longer, just message us). A Super Quest sits beside your other claims and does not count against them.</p>
<p>Marks. Mark the AIS, where the axon leaves the cell; it is the one mark that is required. Others are optional. You could use True End for a branch that stops naturally, and Checked for a place you have looked at and found right, so you know where you have been. Hide marks clears them from the view when you want to see the cell itself. We are working on Highlight mode in MEC. For now, annotations highlight the immediately surrounding area on a cell, so you can add lots of annotations to effectively highlight a branch.</p>
<p>Your time is counted. This is how the lab learns how long these cells really take and will help us estimate the amount of time required for proofreading in this and future datasets. Pyr adds up your active time on the cell and shows it on the row, such as 14 h 20 min so far. It counts only while the cell is on screen, Pyr is the tab in front, and you have used the mouse or keys in the last 5 minutes. There is no need to hurry. It is most important to map the cells accurately.</p>
<p>&quot;Save view&quot; keeps your layers and marks with the claim, and the claim keeps the cell's ID through every merge and split.</p>
<p>A special achievement is waiting for anyone who accurately completes a Super Quest.  Prefer something lighter? <a href="https://connectome.quest/blog/post.html?p=mapping-mec" target="_blank" rel="noopener noreferrer">Try mapping MEC pyramidal axons</a>. Guaranteed only moderately difficult!</p>
<p><strong>Thank you for contributing to neuroscience!</strong></p>]]></content:encoded>
    </item>
    <item>
      <title>How to unlock Edit access in Pyr</title>
      <link>https://connectome.quest/blog/how-to-unlock-edit-access-in-pyr/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/how-to-unlock-edit-access-in-pyr/</guid>
      <pubDate>Fri, 16 Oct 2026 21:08:22 GMT</pubDate>
      <dc:creator>Amy R. Sterling</dc:creator>
      <description>Practice in the Sandbox, then show us ten edits you would make, right inside Pyr. Here is how the access check works, what happens after you send, and where to find help while you learn.</description>
      <content:encoded><![CDATA[<p>The Sandbox is where everyone starts in Pyr. Every signed in player can cut and merge there, and nothing you do in it touches real science. Once those tools feel natural, you can ask for edit access so that you can edit real neurons and help map the brain.</p>
<h2 id="getting-started">Getting started<a class="blog-anchor" href="#getting-started" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Everything you need to learn lives in the menu at the top right of Pyr (the burger icon).</p>
<ul><li><strong>Site Tour.</strong> A quick walk around the screen: the top bar, the views and where everything is.</li><li><strong>Tutorials.</strong> Take them in order. <strong>Get Started</strong> shows you how to move around a cell in 3D and step through the electron microscope images. <strong>Advanced Interface</strong> covers layers, coordinates, annotations and settings. <strong>Merge</strong> and <strong>Cut</strong> teach the two edits on practice cells, so a slip costs nothing.</li><li><strong>Merger Sandbox.</strong> Practice cells, each with a merger to find and cut. The menu keeps count of how many you have done.</li><li><strong>Utilize Chat</strong> if you have questions, or <strong>ask for help in the Cell Library</strong> (neuron icon in the top bar).</li></ul>
<h2 id="what-we-ask-for">What we ask for<a class="blog-anchor" href="#what-we-ask-for" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Before you can contribute to real science, you must demonstrate that you know your way around a neuron. For this, submit test edits on cells in the sandbox or on the Retina dataset where you found mergers or missed extensions. 10 demo edits are currently required:</p>
<ul><li><strong>Seven cuts.</strong> Find places where two cells were wrongly joined. Place red and blue Multicut (cut) points on each side of where you would cut.</li><li><strong>Three merges.</strong> Find places where one cell was wrongly broken in two. Press M, then Ctrl+click the two pieces to place merge points.</li></ul>
<p><strong>Place the points, but do not submit the edit.</strong> Leave each mistake in place, so we can see it the way you found it.</p>
<h2 id="how-to-request-edit-access">How to request Edit access<a class="blog-anchor" href="#how-to-request-edit-access" aria-label="Link to this section" title="Link to this section">#</a></h2>
<ul><li>Open the menu at the top right and choose <strong>Ask for retina access</strong>, at the very bottom.</li><li>Place your cut or merge points, then press <strong>Save this view as cut 1</strong> (or merge 1). Pyr keeps the view, every point you placed, and a note if you want to add one. The points are cleared afterwards, so the next one starts fresh.</li><li>Click any saved example to fly back to it, or press <strong>Replace</strong> to swap in a better one.</li><li>On a laptop, press <strong>Slim</strong> to make the box small while you work.</li><li>When all ten are saved, press <strong>Send for review</strong>.</li></ul>
<h2 id="what-happens-next">What happens next<a class="blog-anchor" href="#what-happens-next" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Someone from the team looks at every example and every point you placed and you will hear back in your notifications.</p>
<p>If your edits look right, your edit access is granted and the world of connectomics opens to you. If a few need another look, we mark them and tell you why. Redo those and send again.</p>
<p>Taking more than one try is normal. You are never locked out and there is no limit on tries. If you have practiced in the Sandbox for two weeks and still feel unsure, reach out to us. We are happy to help.</p>
<h2 id="help-while-you-learn">Help while you learn<a class="blog-anchor" href="#help-while-you-learn" aria-label="Link to this section" title="Link to this section">#</a></h2>
<ul><li><strong>The tutorials in Pyr.</strong> After Get Started, open the menu at the top right and take Merge, Cut and Advanced Interface.</li><li><strong>Chat and the Cell Library.</strong> Ask in chat, or ask for help on a cell from the Cell Library (the neuron icon in the top bar).</li><li><strong>Videos.</strong> <a href="https://www.youtube.com/watch?v=48GS9Sizrvw" target="_blank" rel="noopener noreferrer">How to merge</a>, <a href="https://www.youtube.com/watch?v=DB6wmQWGsck" target="_blank" rel="noopener noreferrer">how to split</a>, and <a href="https://www.youtube.com/watch?v=CGooeAhSryg" target="_blank" rel="noopener noreferrer">Find Path</a>, which helps you trace a merger back to where it starts. There is also a full <a href="https://www.youtube.com/playlist?list=PLZlCbXsRJFCw0BLFWKrc49JHKWK1o41Ud" target="_blank" rel="noopener noreferrer">playlist of advanced tutorials</a>. We are working on tutorials using the new Pyr interface.</li><li><strong>The <a href="https://blog.pyr.ai/2024/12/20/proofreading-101-climb-into-spelunker/" target="_blank" rel="noopener noreferrer">Proofreading Guide</a>.</strong></li><li><strong>Chat</strong></li></ul>
<p>Once you are in, you are part of the team building new maps of the brain. For science!</p>]]></content:encoded>
    </item>
    <item>
      <title>The Cilia Quest: a new way to help in Pyr</title>
      <link>https://connectome.quest/blog/the-cilia-quest-a-new-way-to-help-in-pyr/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/the-cilia-quest-a-new-way-to-help-in-pyr/</guid>
      <pubDate>Tue, 13 Oct 2026 20:45:54 GMT</pubDate>
      <dc:creator>Amy R. Sterling</dc:creator>
      <description>Quests are a new kind of task in Pyr. Your challenge, should you choose to accept it, is to find and tag the primary cilium on starburst cells in the retina, without ever having to proofread the cell.</description>
      <content:encoded><![CDATA[<p>Calling all Pyr players! there's a new <strong>Quest</strong> on the horizon, and we need your help!</p>
<blockquote><p>We're on the hunt for <strong>primary cilia</strong>, important cellular appendages that receive and process environmental signals. Cilia serve as “antennas” that cells use to sense the chemistry of their environment. Each cell only has one primary cilium, will you be the one to find it?</p></blockquote>
<h2 id="how-to-hunt-for-cilia">How to Hunt for Cilia<a class="blog-anchor" href="#how-to-hunt-for-cilia" aria-label="Link to this section" title="Link to this section">#</a></h2>
<ol><li>Open the <strong>Cell Library</strong> and click <strong>Quests</strong>. If you are on another dataset, press <strong>Switch to Retina</strong>.</li><li>Press <strong>Claim</strong> on an available cell. The cell and it's annotation layers will open on screen. Each cell in a Cilia Quest is retinal neuron. Somewhere on its cell body there may be a cilium, a thin appendage that projects from. This is what we quest for.</li><li>Find the cilium (if there is one). At its base sits the basal body, and close by is a second structure, the daughter centriole. Mark all 3 structures, if present.</li><li>Put the crosshairs inside the cell and press <strong>Submit</strong>.</li><li>Choose a cilium status from the list. Don't know which to choose? Select your guess to see help text below. Or <a href="https://connectome.quest/blog/cilia-status-guide/" target="_blank" rel="noopener noreferrer">check out this post</a> for a full explanation of all statuses with examples!</li><li>Select whether the cilium was &quot;split&quot; - part of all of the cilium was not attached, or &quot;merged&quot; - cilium was fully visible in the original segmentation. If no cilium was present, make no selection.</li><li>Add a note if needed, and press the 2nd Submit button!</li></ol>
<p>You can hold three cells at a time. <strong>Release</strong> sends a cell back for someone else.</p>
<figure class="blog-post__figure blog-post__figure--medium" data-size="medium"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/b91ee9ea-336f-47f2-8d53-6b1932cd0398.png" alt="A scientific diagram of the cilium. Check out the Lab's Guidehttps://unitc-my.sharepoint.com/:p:/g/personal/ojxjl01_cloud_uni-tuebingen_de/IQDBsH6sbIjZRY5Gc3Flt5iaAUvXSDf4QLsyamGQ72f8rOo?rtime=Ky2GsGU730g for more about cilia" loading="lazy" /><figcaption>A scientific diagram of the cilium. Check out the <strong><a href="https://unitc-my.sharepoint.com/:p:/g/personal/ojxjl01_cloud_uni-tuebingen_de/IQDBsH6sbIjZRY5Gc3Flt5iaAUvXSDf4QLsyamGQ72f8rOo?rtime=Ky2GsGU730g" target="_blank" rel="noopener noreferrer">Lab's Guide</a></strong> for more about cilia</figcaption></figure>
<p>About each structure:</p>
<ul><li><strong>Cilium</strong> - a columnar structure that projects from the cell's soma. Microtubules should be visible inside (greyish lines).</li><li><strong>Basal body</strong> - a centriole at the base of cilium. It's always inside to the soma membrane. When no cilium is present, there may still be a basal body.</li></ul>
<ul><li><strong>Daughter centriole</strong> - A second centriole. It is often found near, but not attached to the basal body. In some cases it can be very far from the BB -- keep scrolling!</li></ul>
<p><em><strong>Important note!</strong></em> Some cells have no visible cilium, but often you will still be able to locate the two centrioles inside the soma.</p>
<h2 id="annotating-the-structures">annotating the structures<a class="blog-anchor" href="#annotating-the-structures" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>To annotate a cilium, you'll need to use the 3 annotation layers. Click a layer to select it.</p>
<ul><li><strong>Polyline annotation</strong> - use this to annotate the <strong>cilium centerline</strong>. Once selected, Ctrl and click along the cilium, then press Enter. Backspace removes the last point and Esc cancels the line.</li><li><strong>Point annotation</strong> - use this to annotate the <strong>basal body (BB)</strong> and <strong>daughter centriole (DC)</strong>. CTRL + click to place it. Always double check your layer when placing!</li></ul>
<h2 id="putting-it-all-together">Putting it all together<a class="blog-anchor" href="#putting-it-all-together" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5973814101934080" target="_blank" rel="noopener noreferrer">Here's an example</a> of a cell with a cilium that's been fully annotated and is ready to complete!</p>
<figure class="blog-post__figure blog-post__figure--medium" data-size="medium"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/6afa305c257b45ab2518c340/1d7519ca-1702-4af7-8fbb-daa063bb91d4.png" alt="Cell with cilium annotated" loading="lazy" /><figcaption>Cell with cilium annotated</figcaption></figure>
<h2 id="dive-deeper">Dive Deeper<a class="blog-anchor" href="#dive-deeper" aria-label="Link to this section" title="Link to this section">#</a></h2>
<ul><li><a href="https://unitc-my.sharepoint.com/:p:/g/personal/ojxjl01_cloud_uni-tuebingen_de/IQDBsH6sbIjZRY5Gc3Flt5iaAUvXSDf4QLsyamGQ72f8rOo?rtime=Ky2GsGU730g" target="_blank" rel="noopener noreferrer">Check out the lab's scientific guide to cilia with tips on how to locate them.</a></li></ul>
<ul><li><a href="https://www.youtube.com/watch?v=-l2-R-9vX6o&amp;t=1525s" target="_blank" rel="noopener noreferrer">Watch an in-depth training video</a>.</li><li>Here's quick guide to each cilia status and when to use it. Click the image to enlarge, and the link to see an interactive example.</li></ul>
<div class="blog-post__gallery"><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/9f83e1a8-0b69-4dd3-8f54-5d7ea8d523fa.png" alt="Cilium Annotatedhttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6134465776582656 - Cilium emerges from the cell, both centrioles and cilium centerline are annotated." loading="lazy" /><figcaption><strong><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6134465776582656" target="_blank" rel="noopener noreferrer">Cilium Annotated</a></strong> - Cilium emerges from the cell, both centrioles and cilium centerline are annotated.</figcaption></figure><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/94837203-4741-4297-b6f0-aa01f1c472b9.png" alt="Cilium Annotated fixation artifacthttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6224937215852544 - Cilium structure is visible near the surface of the cell, but appears attached/internal to the soma. This is caused by membranes fusing during data collection." loading="lazy" /><figcaption><strong><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6224937215852544" target="_blank" rel="noopener noreferrer">Cilium Annotated (fixation artifact)</a></strong> - Cilium structure is visible near the surface of the cell, but appears attached/internal to the soma. This is caused by membranes fusing during data collection.</figcaption></figure><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/10a74111-9fe5-4e89-b361-e7338a7f7fd3.webp" alt="Need Help: One Centriole Missinghttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6697415730003968 - You found one centriole, but the other is MIA. Cilium may or may not be present" loading="lazy" /><figcaption><strong><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6697415730003968" target="_blank" rel="noopener noreferrer">Need Help: One Centriole Missing</a></strong> - You found one centriole, but the other is MIA. Cilium may or may not be present.</figcaption></figure><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/5bcfbad0-686e-4c3e-8955-055c87b33a4f.png" alt="No Cilium BB &amp; DC clearhttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5742473875816448 - No cilium, but centrioles are visible and clearly differentiated basal body closer to the surface." loading="lazy" /><figcaption><strong><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5742473875816448" target="_blank" rel="noopener noreferrer">No Cilium (BB &amp; DC clear)</a></strong> - No cilium, but centrioles are visible and clearly differentiated (basal body closer to the surface).</figcaption></figure><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/08e39a1d-1f13-4cc7-a278-d2ae50ae302d.png" alt="No cilium BB &amp; DC unclearhttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5185996505219072 - No cilium, centriole positions make it unclear which is the basal body and which is the daughter" loading="lazy" /><figcaption><strong><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/5185996505219072" target="_blank" rel="noopener noreferrer">No cilium (BB &amp; DC unclear)</a></strong> - No cilium, centriole positions make it unclear which is the basal body and which is the daughter.</figcaption></figure><figure class="blog-post__figure"><img src="https://javthknksdcrlhiaaptj.supabase.co/storage/v1/object/public/admin-uploads/blog/69930a0febedc717a52b10fd/c97c0328-eeac-4f1d-9057-c8f7d76ac321.png" alt="Skipped low qualityhttps://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6280972446203904 -- in this case the soma is missing" loading="lazy" /><figcaption><strong><a href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/#!middleauth+https://global.brain-wire-test.org/nglstate/api/v1/6280972446203904" target="_blank" rel="noopener noreferrer">Skipped (low quality)</a></strong> -- in this case the soma is missing.</figcaption></figure></div>
<p>Other statuses:</p>
<ul><li><strong>Need Help</strong> - You can't find any cilium or centrioles, or there's another issue that stops you from annotating the cell. Good to leave clarifying notes in most cases!</li><li><strong>No cilium</strong> - Nothing found. This should only be used after review (use Need Help if you are the first annotator)</li></ul>
<h2 id="where-your-answer-goes">Where your answer goes<a class="blog-anchor" href="#where-your-answer-goes" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>When you submit, your status, your notes and a link to your view go to the lab, with your name and the date. Pyr celebrates with you and tells you how many annotations you placed on that cell and how many Quests you have completed. Your profile keeps the count.</p>
<h2 id="tell-us-how-it-goes">Tell us how it goes<a class="blog-anchor" href="#tell-us-how-it-goes" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>The Cilia Quest is our very first Quest! If something is unclear or could work better, ask in chat or send a report with the feedback button.</p>]]></content:encoded>
    </item>
    <item>
      <title>Achievements in Pyr: how the steps work</title>
      <link>https://connectome.quest/blog/achievements-in-pyr-how-the-steps-work/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/achievements-in-pyr-how-the-steps-work/</guid>
      <pubDate>Sun, 11 Oct 2026 15:23:52 GMT</pubDate>
      <dc:creator>Amy R. Sterling</dc:creator>
      <description>Overview of achievement tiers in Pyr.</description>
      <content:encoded><![CDATA[<p>Pyr has two sets of Achievements you earn just by doing science. One counts the <strong>cells you complete</strong>. The other counts the <strong>edits you make</strong>.</p>
<p>Achievements you earned before the November 2026 update stay yours, in the Legacy Achievements row of your Trophy Case.</p>
<p>Below find a list of the thresholds.</p>
<h2 id="cell-completion-achievements">Cell Completion Achievements<a class="blog-anchor" href="#cell-completion-achievements" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Every cell you complete counts.</p>
<ul><li>Your <strong>first cell</strong> earns your first Achievement.</li></ul>
<ul><li>Then <strong>5, 10, 15 and 20</strong> cells.</li></ul>
<ul><li>Then <strong>30, 40, 50, 75, 100, 125 and 150</strong> cells.</li></ul>
<ul><li>Every <strong>50</strong> cells up to 500.</li></ul>
<ul><li>Every <strong>100</strong> cells up to 2,000.</li></ul>
<ul><li>Every <strong>250</strong> cells up to 5,000.</li></ul>
<ul><li>Every <strong>500</strong> cells up to 10,000.</li></ul>
<ul><li>After 10,000, Every <strong>1,000</strong> cells.</li></ul>
<h2 id="editor-achievements">Editor Achievements<a class="blog-anchor" href="#editor-achievements" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Every merge and every split counts as an edit.</p>
<ul><li>Your <strong>first edit</strong> earns your first Achievement.</li></ul>
<ul><li>Then <strong>10, 25, 50 and 100</strong> edits.</li></ul>
<ul><li>Every <strong>100</strong> edits up to 500.</li></ul>
<ul><li>Then <strong>750 and 1,000</strong> edits.</li></ul>
<ul><li>Every <strong>500</strong> edits up to 5,000.</li></ul>
<ul><li>Every <strong>1,000</strong> edits up to 20,000.</li></ul>
<ul><li>Every <strong>2,500</strong> edits up to 50,000.</li></ul>
<ul><li>Every <strong>5,000</strong> edits up to 100,000.</li></ul>
<ul><li>After 100,000, every <strong>25,000</strong> edits.</li></ul>
<h2 id="what-will-you-unlock-next">What will you unlock next?<a class="blog-anchor" href="#what-will-you-unlock-next" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Open your Trophy Case to see the ones you hold and how far you are from the next one.</p>
<p>Thank you for helping to map the brain. For science!</p>]]></content:encoded>
    </item>
    <item>
      <title>Upgrades to chat</title>
      <link>https://connectome.quest/blog/upgrades-to-chat/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/upgrades-to-chat/</guid>
      <pubDate>Sat, 10 Oct 2026 16:46:31 GMT</pubDate>
      <dc:creator>Amy R. Sterling</dc:creator>
      <description>Chat in Pyr has a set of new tools. Search the conversation, add bold and underline, see where new messages begin, and find your own messages at a glance.</description>
      <content:encoded><![CDATA[<p>Chat is where a lot of Pyr happens: questions about a tricky branch, a link to a cell, a cheer when someone finishes a big one or earns a new achievement. This week it picked up a set of expansions suggested by Krzysztof Kruk. Bonus: None of them takes up any more of your screen.</p>
<h2 id="bold-and-underline">Bold and underline<a class="blog-anchor" href="#bold-and-underline" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>There is no toolbar to learn. Put a star, <code>*</code>, on each side of the words you want in bold. Put an underscore on each side of the words you want underlined, like <code>_this_</code>.</p>
<p>Or select some text in the message box and press <strong>Ctrl+B</strong> for bold or <strong>Ctrl+U</strong> for underline. Press it again to take the mark off. You can use both on the same words.</p>
<p>The marks only count around a whole word or phrase. A name with an underscore in it, or a sum written with stars, stays exactly as you typed it.</p>
<h2 id="search-the-conversation">Search the conversation<a class="blog-anchor" href="#search-the-conversation" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Click the magnifier at the top of the chat. The line where you type a message becomes a search line, and the chat shows only the messages that match.</p>
<ul><li>Type one or more words. A message has to contain all of them.</li></ul>
<ul><li>Put a minus sign in front of a word to leave out messages that have it, like <code>merge -queue</code>.</li></ul>
<ul><li>Put words in quotes to find them together, like <code>&quot;cut off&quot;</code>.</li></ul>
<ul><li>Type <code>from:</code> and a name to see one player's messages, like <code>from:Nseraf</code>.</li></ul>
<ul><li>Search one day with <code>on:10/8</code>, or a stretch of days with <code>after:yesterday</code> and <code>before:2026-10-01</code>. The words today and yesterday work too.</li></ul>
<p>Click any result to jump to that message in the conversation, with everything around it. Press <strong>Esc</strong>, or click the magnifier again, to go back to chat.</p>
<p>Search looks through the messages chat has loaded. To reach further back, press <strong>Search earlier messages too</strong> at the top of the list to include the past 14 days. When you search for a day, chat brings in the earlier messages by itself.</p>
<h2 id="find-your-own-messages">Find your own messages<a class="blog-anchor" href="#find-your-own-messages" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>A thin bar now appears to the left of your own messages, making them easy to spot. If you would rather not see it, open <strong>Settings</strong> and switch off <strong>Mark my own messages in chat</strong>.</p>
<h2 id="see-what-is-new">See what is new<a class="blog-anchor" href="#see-what-is-new" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>When you open Pyr, or come back to it from another tab, a line across the chat shows where the messages you have not seen begin. It reads <strong>New since you left</strong>. Everything below it is new. The line goes away once you send a message.</p>
<h2 id="pinned-messages">Pinned messages<a class="blog-anchor" href="#pinned-messages" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Sometimes there is one thing everyone should see, like a dataset opening or an event starting. The Pyr team can now pin a message to the top of chat. It sits in one line above the conversation. Click it to jump to the message, or press the small x to hide it for yourself.</p>
<h2 id="thank-you-krzysztof">Thank you, Krzysztof<a class="blog-anchor" href="#thank-you-krzysztof" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Search, bold and underline, and marking your own messages all came from one player's suggestions. If you have an idea for Pyr, send it in with the feedback button.   For science!</p>]]></content:encoded>
    </item>
    <item>
      <title>Highlight mode: color in what you have checked</title>
      <link>https://connectome.quest/blog/highlight-mode/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/highlight-mode/</guid>
      <pubDate>Mon, 05 Oct 2026 23:57:46 GMT</pubDate>
      <dc:creator>Amy R. Sterling</dc:creator>
      <description>A new tool for keeping track of your work on a big cell. Mark a stretch of a branch and its surface takes a color, so you can see what is done and what is left.</description>
      <content:encoded><![CDATA[<p>Proofreading a big cell means checking a lot of branches, and it is easy to lose track of which ones you have already been down. Until now the usual trick was to drop a point at the end of each branch. That marks a place. It does not show you the stretch of cell you checked.</p>
<p>Highlight mode does. You pick two points on a cell, and the stretch between them takes a color. The cell fills in as you work, and whatever is still plain is what is left to do.</p>
<h2 id="how-to-use-it">How to use it<a class="blog-anchor" href="#how-to-use-it" aria-label="Link to this section" title="Link to this section">#</a></h2>
<ol><li>Click the spray can in the toolbar to open Highlight.</li><li>Hold <strong>Ctrl</strong> and click the cell where your checked stretch starts. On a Mac, hold <strong>Cmd</strong>. Holding <strong>H</strong> and clicking works too.</li><li>Turn and move the view as much as you like.</li><li>Hold <strong>Ctrl</strong> and click where the stretch ends.</li></ol>
<p>The game finds the path along the cell between your two clicks and colors the surface of the cell along it.</p>
<h2 id="pick-your-colors">Pick your colors<a class="blog-anchor" href="#pick-your-colors" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Highlight starts with three colors: <strong>Checked</strong>, <strong>Needs a look</strong> and <strong>Problem</strong>. Click a color's swatch to change it. Use <strong>+ Add</strong> to make your own, with any name you like, up to eight colors in all.</p>
<p>Your colors are saved to your account, so they are there on any computer you sign in on.</p>
<h2 id="color-reach">Color reach<a class="blog-anchor" href="#color-reach" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>The <strong>Color reach</strong> slider sets how far out from the middle of the branch the color spreads, from 0.5 to 8 µm.</p>
<p>Here is the same stretch of dendrite at two settings. At 2.5 µm the color covers the shaft, but the tips of the longer spines stay the cell's own purple.</p>
<figure class="blog-post__figure"><img src="https://connectome.quest/assets/images/blog/highlight-reach-2-5-um.webp" alt="A dendrite highlighted blue with color reach at 2.5 micrometers. The tips of the longer spines are still purple." loading="lazy" /></figure>
<p>At 4 µm the color reaches all the way to the spine tips.</p>
<figure class="blog-post__figure"><img src="https://connectome.quest/assets/images/blog/highlight-reach-4-um.webp" alt="The same dendrite with color reach at 4 micrometers. The spines are blue to their tips." loading="lazy" /></figure>
<p>A wider reach covers thick branches and their spines. A narrower reach keeps the color off neighboring branches that run close by. You can move the slider after you have made your marks, and they all update.</p>
<h2 id="good-to-know">Good to know<a class="blog-anchor" href="#good-to-know" aria-label="Link to this section" title="Link to this section">#</a></h2>
<ul><li><strong>Undo</strong> removes your last mark. <strong>Clear all</strong> removes every mark.</li><li><strong>Your marks are saved with your view.</strong> They come back with the autosave, with Save view on a claim, and in a link you share.</li><li><strong>Marks stay put when you merge or split.</strong> They are attached to the place, so editing the cell does not lose them.</li><li><strong>Long paths can take a while.</strong> If the game cannot find the path within a minute it stops, and nothing is marked. There is also a <strong>Stop tracing</strong> button.</li><li><strong>You can drag the Highlight box</strong> anywhere on the screen, and it stays where you put it.</li><li><strong>Highlight is not on every dataset yet.</strong> Where it is off, the spray can is greyed out.</li></ul>
<h2 id="tell-us-how-it-goes">Tell us how it goes<a class="blog-anchor" href="#tell-us-how-it-goes" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Highlight is new, and it will change as you use it. If something does not color the way you expect, send it with the feedback button in the game, or <a href="https://connectome.quest/forum.html" target="_blank" rel="noopener noreferrer">tell us on the forum</a>.</p>]]></content:encoded>
    </item>
    <item>
      <title>Write your own scripts for EyeWire II</title>
      <link>https://connectome.quest/blog/scripts-and-api/</link>
      <guid isPermaLink="true">https://connectome.quest/blog/scripts-and-api/</guid>
      <pubDate>Fri, 02 Oct 2026 16:00:00 GMT</pubDate>
      <dc:creator>Amy and the EyeWire team</dc:creator>
      <description>EyeWire II now has a small set of tools for player scripts, plus a guide to get you started with Tampermonkey. Also this week: split screen replaces the 4 panel view, your view saves itself, and the fly nerve cord joins the dataset list.</description>
      <content:encoded><![CDATA[<p>Players of the original EyeWire built some of its best features themselves. Bigger annotation dots, extra statistics, faster ways to get around: many of them started as a script one player wrote for their own browser and then shared.</p>
<p>EyeWire II now supports that properly. There is a guide for getting started, and a small set of tools made for scripts to use.</p>
<h2 id="what-a-script-is">What a script is<a class="blog-anchor" href="#what-a-script-is" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>A script is a small piece of code that changes the game on your computer only. It runs through Tampermonkey, a free browser extension. Nobody else sees your script, and you do not need permission to write one.</p>
<h2 id="what-scripts-can-use">What scripts can use<a class="blog-anchor" href="#what-scripts-can-use" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>Until now a script had to reach into the inner workings of the game, which change with almost every update. So scripts broke often. The new tools are the parts we promise to keep working:</p>
<ul><li><strong>Your cells and your view.</strong> The cells you have selected, the cell under the crosshairs, your claims, and where you are looking.</li><li><strong>Events.</strong> Run your code when you claim, release or complete a cell, when your selection changes, or when you switch datasets.</li><li><strong>A button and a panel.</strong> Add your own button to the top bar and your own side panel.</li><li><strong>Annotation layers.</strong> Make a layer of your own and add points and lines to it.</li><li><strong>Settings.</strong> Save your script's settings in your browser.</li></ul>
<p>Everything else is still open. Scripts can reach the viewer directly for anything the tools do not cover.</p>
<p class="blog-post__cta"><a class="blog-post__button" href="https://eyewire-ii-community-dot-brain-wire-dot-seung-lab.ue.r.appspot.com/scripts.html" target="_blank" rel="noopener noreferrer">Open the scripts guide</a></p>
<p>The guide has setup steps, the full list of tools, and six example scripts you can copy.</p>
<h2 id="a-word-on-safety">A word on safety<a class="blog-anchor" href="#a-word-on-safety" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>A script runs as you. It can do anything you can do in the game, including editing cells with your sign in. Scripts you write yourself are fine. Before you install a script from someone else, read it, or ask in chat whether others use it. Scripts are made by players and are not reviewed by the EyeWire team.</p>
<h2 id="tell-us-what-is-missing">Tell us what is missing<a class="blog-anchor" href="#tell-us-what-is-missing" aria-label="Link to this section" title="Link to this section">#</a></h2>
<p>We asked the script writers on the forum what they need, and the answer was that scripts grow out of whatever rough edge someone notices that week. So the list of tools starts small on purpose. If your script needs something that is not there, <a href="https://forum.eyewire.org/t/pyr-eyewire-ii-api-requests/2246/4" target="_blank" rel="noopener noreferrer">ask on the forum</a> and we will add it.</p>
<h2 id="also-new-this-week">Also new this week<a class="blog-anchor" href="#also-new-this-week" aria-label="Link to this section" title="Link to this section">#</a></h2>
<ul><li><strong>Split screen replaces the 4 panel view.</strong> Press Space to flip between split screen and the single panel under your mouse. Nothing lands you in 4 panel any more.</li><li><strong>Your view saves itself.</strong> Annotations, layers and position are saved to your account as you work. Open the game on any computer and it offers to pick up where you left off.</li><li><strong>Save view on a claim.</strong> In the Cell Library, the cell you are working on has a Save view button, so your annotations stay with the cell.</li><li><strong>The fly brain and nerve cord.</strong> BANC is now in the dataset switcher, and the three MICrONS versions share one card.</li><li><strong>Claim several cells at once.</strong> The Cell Library can fill your claims from the Available list in one click.</li><li><strong>A faster Cell Library.</strong> Your own cells load first, and long lists draw 200 rows at a time.</li></ul>]]></content:encoded>
    </item>
  </channel>
</rss>
