Confocal analysis, tissue tracking, and figures in one neuroscience workspace
A neuroscience result is a count, an intensity, or a colocalization, traced back to an animal and a brain region. Conspecta keeps the z-stack, the measurement, the section it came from, and the figure it ends up in inside a single project.
The neuroscience imaging bench, in one browser tab
One project. Every stack, count, and figure stays linked to the animal, brain region, section, and injection site behind it.
Import
Upload LIF, TIFF, PNG, and JPEG into the project the tissue already lives in. Conspecta starts after acquisition and does not drive the scope.
How most labs do it today. The stacks land on the rig computer in a shared core, then leave on an external drive. The first copy of a two-week imaging run is often the only copy.
Z-stacks
Multi-channel z-stacks open directly in the browser, with independent brightness and pseudocolor per channel, plane-by-plane browsing, and maximum intensity, average, or sum projections.
How most labs do it today. A four-channel, sixty-plane stack opens on the workstation that has the right Java and Bio-Formats build, and nowhere else. Nothing opens on the student's managed laptop.
Counting
Open a stack and a segmentation model counts every neuron, with no threshold to tune. Count on the projection you're viewing, or per plane across the stack for a depth profile.
How most labs do it today. A threshold that separates puncta in a bright section clips them in a dim one, so it gets nudged per image, and the value behind a count isn't stored with the count. The macro that did it lives in someone's home directory.
Colocalization
Pearson's correlation, Manders' coefficients, and Costes thresholding, plus per-cell marker intensity across channels with positive and negative classification, each measurement still attached to the section it came from.
How most labs do it today. The result is a number in a log window. Which section, which channel pair, and which animal produced it lives in the filename, if anywhere.
Plots
Measurements feed charts directly, grouped by animal, region, and condition. Re-run a detection and the numbers downstream follow it.
How most labs do it today. Re-running the quantification means a fresh export and a fresh paste, and the plot keeps no record of which measurement produced it.
Animals
Track animals, brain regions, sections, and injection sites as samples, and every image and measurement stays linked to the tissue it came from.
How most labs do it today. Which stack was animal 7, contralateral hemisphere, third section from the injection site? The answer is a filename convention that one rename destroys.
Writing
Notes, image analysis, and multi-panel figures sit in one project, with panel labels, and the figure is built from the analysis rather than pasted beside it.
How most labs do it today. The notebook describes the quantification in prose. The figure is a screenshot of an analysis that has since been re-run twice.
See it on your own stack
Upload a confocal z-stack and run a count. Every number stays attached to the image and the channel it was measured from.