cellbrowser 1.2.16__tar.gz → 1.2.17__tar.gz

{cellbrowser-1.2.16/src/cbPyLib/cellbrowser.egg-info → cellbrowser-1.2.17}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cellbrowser
-Version: 1.2.16
+Version: 1.2.17
 Summary: UCSC Cellbrowser, an interactive browser for single cell data. Includes converters and basic pipelines for text files, Seurat, Scanpy and Cellranger.
 Home-page: https://github.com/maximilianh/cellBrowser
 Author: Maximilian Haeussler

{cellbrowser-1.2.16 → cellbrowser-1.2.17}/README.rst

@@ -14,29 +14,36 @@ you can always use tsv/csv files to import your data. The exporters are part of
 
 Here are a few datasets that demonstrate these features:
 
-- A simple gene expression dataset: https://cells.ucsc.edu/?ds=cortex-dev
+- A simple gene expression dataset: https://cortex-dev.cells.ucsc.edu
+- ATAC support https://cortex-atac.cells.ucsc.edu
 - Spatial transcriptomics support https://ms-subcortical-lesions.cells.ucsc.edu
 - Split screen mode to display both the spatial and snRNA-seq data side by side https://cells.ucsc.edu/?ds=dup15q-cortex-organoids+spatial+control
-- Cell trajectories from monocle https://cells-test.gi.ucsc.edu/?ds=pre-postnatal-cortex+ex-neu+rna
-- Cell trajectories from URD https://cells.ucsc.edu/?ds=cardiac-differentiation+trajectory+cm-combined-trajectory
+- Pseudotime cell trajectories from Monocle3 https://cells-test.gi.ucsc.edu/?ds=pre-postnatal-cortex+ex-neu+rna
+- Tree-like cell trajectories from URD https://cells.ucsc.edu/?ds=cardiac-differentiation+trajectory+cm-combined-trajectory
 - Clone lineage tracing support: color by the field CellTag, select tags in the legend, then click "recolor checked": https://cells.ucsc.edu/?ds=gbm-nvp+nvp-celltag
 - Brain lipidomics https://cells.ucsc.edu/?ds=brain-lipids
 
-To show all our > 200 single cell datasets, see http://cells.ucsc.edu
+To show all our > 300 single cell datasets, see http://cells.ucsc.edu
 
-To setup your own cell browser, from Cellranger, Seurat, Scanpy or text files 
+To upload datasets for the cells.ucsc.edu website, use https://cells-submit.gi.ucsc.edu. Or email us at cells@ucsc.edu. We reply usually on the same day.
+
+Most users prefer uploading their datasets to us rather than hosting them on their own webserver.
+Our webserver is reliable, fast and we take care
+of adding new features and keeping the data online. 
+But, if you really want to make your own website, to setup your own cell
+browser, on your own web server, from Cellranger, Seurat, Scanpy or text files
 (tsv/csv), or just a single cell expression matrix, read the documentation
-at http://cellbrowser.rtfd.io. If you use the UCSC Cell Browser in your research, please cite
+at http://cellbrowser.rtfd.io.
+
+If you use the UCSC Cell Browser in your research, please cite
 `our Bioinformatics paper <https://dx.doi.org/10.1093/bioinformatics/btab503>`_.
 If you are also using data from a specific dataset we host, please also cite
 the original authors of that dataset (visible under 'Info & Download' while viewing that dataset).
 
-If you want us to add a single cell dataset to the website http://cells.ucsc.edu, 
-please contact us at cells@ucsc.edu. We are happy to add any dataset.
-
 This is a viewer for a static, precomputed layout. If you're looking for an interative layout, where you can 
 move the cells around and run some algorithms interactively, try Chan-Zuckerberg's own cellxgene or Spring.
-A website with both datasets and some analysis is `Scope <http://scope.aertslab.org/>`_.
+Another website with both datasets and some analysis is `Scope <http://scope.aertslab.org/>`_. There are many other
+similar websites now, usually with a few dozen datasets.
 
 Many labs host their data at cells.ucsc.edu by sending it to us, but some groups have setup their own cell browsers:
 
@@ -58,8 +65,7 @@ Many labs host their data at cells.ucsc.edu by sending it to us, but some groups
 * Paul Gontarz, WUSTL, http://regmedsrv1.wustl.edu/Public_SPACE/pgontarz/Public_html/cellbrower/Exp1/
 * Dylan Farnsworth, U Oregon, https://pages.uoregon.edu/drf/browser/lens_230620/
 
-
-These papers have cell browsers made at UCSC:
+These are some papers that reference cell browsers made at UCSC but there are many more, search for "cells.ucsc.edu" in Google Scholar to find more:
 
 * organoidatlas: https://www.sciencedirect.com/science/article/pii/S221112472030053X
 * dros-brain: https://elifesciences.org/articles/50354
@@ -69,8 +75,9 @@ These papers have cell browsers made at UCSC:
 
 Before judging this project by the number of issue tickets or PRs, note that at UCSC we use an internal
 ticket system with more features and that a lot of communication with wetlab users is by email at cells@ucsc.edu, as we 
-do not require a Github account for feedback. But we do reply to issues here, as you can see from the Github 
-account and also use Github for source control.
+do not require a Github account for feedback, as our submitters are biologists, not computer scientists.
+But we do reply to issues here, as you can see from the Github account and also use Github for source control, and 
+we get emails every day. The project is much more active than the Github issues suggest.
 
 Install and use
 ---------------
@@ -82,9 +89,9 @@ Install and use
 * Galaxy: there is a Galaxy tool for UCSC CellBrowser, which can be installed on any Galaxy instance via its `Galaxy Toolshed entry <https://toolshed.g2.bx.psu.edu/view/ebi-gxa/ucsc_cell_browser>`_ or it can be directly used by users at the `Human Cell Atlas Galaxy instance <https://humancellatlas.usegalaxy.eu/root?tool_id=toolshed.g2.bx.psu.edu/repos/ebi-gxa/ucsc_cell_browser/ucsc_cell_browser>`_ or as part of the example workflows, such as the `Human Cell Atlas / Scanpy CellBrowser workflow <https://humancellatlas.usegalaxy.eu/u/pmoreno/w/humancellatlas-scanpy-cellbrowser>`_ or the `EBI Single Cell Expression Atlas / Scanpy / CellBrowser workflow <https://humancellatlas.usegalaxy.eu/u/pmoreno/w/atlas-scanpy-cellbrowser-imported-from-uploaded-file>`_
 
 This project was funded by the California Institute of Regenerative Medicine and the
-Chan-Zuckerberg Initiative https://www.chanzuckerberg.com/. In 2020, it was funded through a supplement to the NHGRI Genome Browser grant. Since 2023, it is funded by a grant from NIMH BRAIN.
+Chan-Zuckerberg Initiative https://www.chanzuckerberg.com/. In 2020, it was funded through a supplement to the NHGRI Genome Browser grant. Since 2023, it is funded by a grant from NIMH BRAIN and a DISC0 from CIRM.
 
-This is early research software. You are likely to find bugs. Please open a Github
+This is early research software. It may contain bugs. Please open a Github
 ticket or email us at cells@ucsc.edu, we can usually fix them quickly.
 
 Citation

{cellbrowser-1.2.16 → cellbrowser-1.2.17}/src/cbPyLib/cellbrowser/_version.py

@@ -8,11 +8,11 @@ import json
 
 version_json = '''
 {
- "date": "2025-11-12T13:17:35-0800",
+ "date": "2026-04-22T13:09:32-0700",
  "dirty": false,
  "error": null,
- "full-revisionid": "9d80bedf1b39a0e74cf2e6f9f1481d6b6a66c9cc",
- "version": "1.2.16"
+ "full-revisionid": "73100b2c3973b545756150bc92f27113edf0f592",
+ "version": "1.2.17"
 }
 '''  # END VERSION_JSON
 

{cellbrowser-1.2.16 → cellbrowser-1.2.17}/src/cbPyLib/cellbrowser/cbWeb/js/cellBrowser.js

@@ -768,8 +768,7 @@ var cellbrowser = function() {
                 htmls.push("select a dataset in the list to the left, and navigate to the 'Data Download' tab. ");
                 htmls.push("This information can also be accessed while viewing a dataset by clicking the 'Info & Downloads' button.");
             } else if (desc.hideDownload===true || desc.hideDownload=="True" || desc.hideDownload=="true") {
-                htmls.push("The downloads section has been deactivated by the authors.");
-                htmls.push("Please contact the dataset authors to get access.");
+                htmls.push("The downloads section has been deactivated by the authors. Please contact the dataset authors to get access.");
             } else {
                 if (desc.matrices) {
                     htmls.push("<p>");
@@ -2882,7 +2881,8 @@ var cellbrowser = function() {
          htmls.push('<li><a href="#" id="tpAboutButton">About</a></li>');
          htmls.push('<li><a href="https://cellbrowser.readthedocs.io/en/master/interface.html" target=_blank id="tpQuickstartButton">How to use this website</a></li>');
          htmls.push('<li><a href="#" id="tpTutorialButton">Interactive Tutorial</a></li>');
-         htmls.push('<li><a target=_blank href="https://github.com/maximilianh/cellBrowser#readme" id="tpGithubButton">Setup your own cell browser</a></li>');
+         htmls.push('<li><a target=_blank href="https://cells-submit.gi.ucsc.edu" id="tpSubmitButton">Upload your own data</a></li>');
+         htmls.push('<li><a target=_blank href="https://github.com/ucscGenomeBrowser/cellBrowser#readme" id="tpGithubButton">Setup your own cell browser</a></li>');
          htmls.push('</ul>'); // Help dropdown-menu
          htmls.push('</li>'); // Help dropdown container
 
@@ -9807,6 +9807,8 @@ function onClusterNameHover(clusterName, nameIdx, ev, isLegend) {
         buildMenuBar();
 
         var datasetName = getDatasetNameFromUrl()
+        if (datasetName === "whole-brain-perturb" && window.location.hostname === "cells.ucsc.edu")
+            window.location.replace("https://cells-test.gi.ucsc.edu/?ds=" + datasetName);
         // pre-load dataset.json here?
         menuBarHeight = $('#tpMenuBar').outerHeight(true);
 

cellbrowser-1.2.17/src/cbPyLib/cellbrowser/cbWeb/js/violinWorker.js

@@ -0,0 +1,134 @@
+/*
+ * violinWorker.js - Web Worker for violin plot statistics computation.
+ *
+ * Receives Float32Array expression vectors from the main thread via zero-copy
+ * Transferable, computes violin statistics off the main thread, and posts back
+ * compact summary data (quantiles + KDE density curve) for rendering.
+ *
+ * Message in:  { arrays: [Float32Array, ...], labels: [...], doLog2: bool, reqId: int }
+ * Message out: { results: [{...stats...}, ...], labels: [...], reqId: int }
+ */
+
+"use strict";
+
+self.onmessage = function(e) {
+    var arrays  = e.data.arrays;
+    var labels  = e.data.labels;
+    var doLog2  = e.data.doLog2;
+    var reqId   = e.data.reqId;
+
+    var results = [];
+    for (var i = 0; i < arrays.length; i++) {
+        results.push(computeViolinStats(arrays[i], doLog2));
+    }
+
+    self.postMessage({ results: results, labels: labels, reqId: reqId });
+};
+
+function log2Transform(arr) {
+    /* apply log2(x + 1) in-place */
+    for (var i = 0; i < arr.length; i++) {
+        arr[i] = Math.log2(arr[i] + 1);
+    }
+}
+
+function computeViolinStats(arr, doLog2) {
+    var n = arr.length;
+    if (n === 0) {
+        return { q1:0, q3:0, median:0, whiskerLow:0, whiskerHigh:0,
+                 minVal:0, maxVal:0, densityX:new Float32Array(0),
+                 densityY:new Float32Array(0), n:0 };
+    }
+
+    if (doLog2) {
+        log2Transform(arr);
+    }
+
+    /* Sort in-place. TypedArray.sort() uses a native comparison (no toString
+     * overhead), and runs off the main thread so the UI stays responsive. */
+    arr.sort();
+
+    var minVal = arr[0];
+    var maxVal = arr[n - 1];
+
+    /* Quantiles via index into the sorted array */
+    var q1     = arr[Math.floor(n * 0.25)];
+    var median = arr[Math.floor(n * 0.50)];
+    var q3     = arr[Math.floor(n * 0.75)];
+    var iqr    = q3 - q1;
+
+    /* Tukey fences for whiskers */
+    var fenceLow  = q1 - 1.5 * iqr;
+    var fenceHigh = q3 + 1.5 * iqr;
+
+    /* Walk sorted array to find actual data points within fences */
+    var whiskerLow  = minVal;
+    var whiskerHigh = maxVal;
+    for (var i = 0; i < n; i++) {
+        if (arr[i] >= fenceLow)  { whiskerLow  = arr[i]; break; }
+    }
+    for (var i = n - 1; i >= 0; i--) {
+        if (arr[i] <= fenceHigh) { whiskerHigh = arr[i]; break; }
+    }
+
+    /* Edge case: all values are identical */
+    var range = maxVal - minVal;
+    if (range === 0) {
+        var dX = new Float32Array([minVal]);
+        var dY = new Float32Array([1]);
+        return { q1, q3, median, whiskerLow, whiskerHigh, minVal, maxVal,
+                 densityX: dX, densityY: dY, n };
+    }
+
+    /* ---- Histogram-based KDE ------------------------------------------ *
+     * One O(n) pass to fill 512 bins, then convolve with a Gaussian kernel.
+     * This gives a smooth density estimate without evaluating a kernel n
+     * times per query point (which would be O(n * nPoints)).
+     * -------------------------------------------------------------------- */
+    var nBins = 512;
+    var hist  = new Float64Array(nBins);
+    var scale = nBins / range;
+
+    for (var i = 0; i < n; i++) {
+        var bin = Math.min(nBins - 1, Math.floor((arr[i] - minVal) * scale));
+        hist[bin]++;
+    }
+
+    /* Gaussian kernel: sigma chosen so the smooth bandwidth is ~3% of range */
+    var sigma   = nBins / 30;
+    var kRadius = Math.ceil(3 * sigma);
+    var kernel  = new Float64Array(2 * kRadius + 1);
+    var kSum    = 0;
+    for (var k = -kRadius; k <= kRadius; k++) {
+        var kv = Math.exp(-0.5 * k * k / (sigma * sigma));
+        kernel[k + kRadius] = kv;
+        kSum += kv;
+    }
+    for (var k = 0; k < kernel.length; k++) kernel[k] /= kSum;
+
+    var smoothed = new Float64Array(nBins);
+    for (var i = 0; i < nBins; i++) {
+        var val = 0;
+        for (var k = -kRadius; k <= kRadius; k++) {
+            var bi = i + k;
+            if (bi >= 0 && bi < nBins) val += hist[bi] * kernel[k + kRadius];
+        }
+        smoothed[i] = val;
+    }
+
+    /* Normalise density to [0, 1] */
+    var smoothMax = 0;
+    for (var i = 0; i < nBins; i++) {
+        if (smoothed[i] > smoothMax) smoothMax = smoothed[i];
+    }
+
+    var densityX = new Float32Array(nBins);
+    var densityY = new Float32Array(nBins);
+    for (var i = 0; i < nBins; i++) {
+        densityX[i] = minVal + (i / (nBins - 1)) * range;
+        densityY[i] = smoothed[i] / smoothMax;
+    }
+
+    return { q1, q3, median, whiskerLow, whiskerHigh, minVal, maxVal,
+             densityX, densityY, n };
+}

{cellbrowser-1.2.16 → cellbrowser-1.2.17/src/cbPyLib/cellbrowser.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cellbrowser
-Version: 1.2.16
+Version: 1.2.17
 Summary: UCSC Cellbrowser, an interactive browser for single cell data. Includes converters and basic pipelines for text files, Seurat, Scanpy and Cellranger.
 Home-page: https://github.com/maximilianh/cellBrowser
 Author: Maximilian Haeussler

{cellbrowser-1.2.16 → cellbrowser-1.2.17}/src/cbPyLib/cellbrowser.egg-info/SOURCES.txt

@@ -171,6 +171,7 @@ src/cbPyLib/cellbrowser/cbWeb/js/cbData.js
 src/cbPyLib/cellbrowser/cbWeb/js/cellBrowser.js
 src/cbPyLib/cellbrowser/cbWeb/js/maxHeat.js
 src/cbPyLib/cellbrowser/cbWeb/js/maxPlot.js
+src/cbPyLib/cellbrowser/cbWeb/js/violinWorker.js
 src/cbPyLib/cellbrowser/sampleConfig/cellbrowser.conf
 src/cbPyLib/cellbrowser/sampleConfig/desc.conf
 src/cbPyLib/cellbrowser/sampleConfig/hub.conf