Example: a Plate with Analysis Results
Most examples show pixels. This one shows what an analysis adds on top: masks for every nucleus and cell, and a measurement table behind each of them. It opens in one click, straight from our site.
Open the example in the viewer
Or open the Example Gallery and pick BBBC013 FKHR translocation. The Analytical Results filter shows only examples that come with results.
What is in it
The plate. BBBC013 from the Broad Bioimage Benchmark Collection. Human U2OS cells express FKHR-GFP, a protein that moves from the cytoplasm into the nucleus when the PI3K pathway is blocked. Two inhibitors run as dose series across the plate: Wortmannin in rows A to D, LY294002 in rows E to H. Two channels: DRAQ for DNA, and FKHR-GFP.
The results. One analysis run, three compartments:
| Mask | Objects | Made with |
|---|---|---|
| nuclei | 18,648 | StarDist (2D_versatile_fluo) |
| cells | 15,331 | Cellpose (cyto3) |
| cytoplasm | 15,272 | each cell minus its nucleus |
Each mask carries its own table: area, shape and intensity for every object, plus well and plate summaries.
Look around
- The whole plate opens inverted on white. Cells are coloured by the phenotype the analysis gave them: positive, negative or excluded. Nuclei are drawn in grey on top.
- The Labels tab lists the three masks under one heading, the analysis run. Press the
on a mask to show or hide it. - Click the nuclei row to open its details. The Analytics section lists the measured columns.
- Colour the nuclei by a column, filter them by value, or build a phenotype. See Analytics & Phenotyping.
- Zoom into a well, or pick one from the Navigation tab. See HCS Plates.
A good first column: colour the nuclei by ch1_mean, the mean FKHR-GFP inside each nucleus. In wells with more inhibitor, more of the signal sits in the nucleus.
Who made the masks
We ran the Find Nuclei Analysis Tool over the plate. The masks and measurements are our output, not part of the original collection.
Use your own results
Everything here is plain files on a web server: an OME-Zarr plate, and next to it a folder with masks, Parquet tables and a manifest. Put yours on any host that answers the browser directly, and the same link pattern opens them:
?url=<plate>&analyticsUrl=<run-folder>/The Analytics Data Format describes the files, from a single CSV up to a full run. Sharing Views covers what else a link can carry.
For a walkthrough of the analysis itself on this same screen, read HCS Analytics in the Browser: 18,648 Nuclei, Zero Uploads.
Image credits
Images: BBBC013v1, provided by Ilya Ravkin, available from the Broad Bioimage Benchmark Collection, under CC BY 3.0.
Ljosa V, Sokolnicki KL, Carpenter AE (2012). Annotated high-throughput microscopy image sets for validation. Nature Methods 9(7):637. doi:10.1038/nmeth.2083
See Sample Data and Credits for every dataset used in these guides.