Mockup 10 · evidence card · the embedded-method unit

One card: the result first, the method folded in

Kate's note from Jun 10: "I'd never think about evidence and method separately" — so they're not separate. The Evidence card is the unit: result and plot on top, method collapsed inline, full study details one click deeper. Below, the same card at four states.

🔒 mira.science/n/sg-composition-019ea6d6 ⌘ L
MIRA · evidence card · spec
Stress-granule formation · Vogel × Kate
K viewing as Kate
Collapsed (PI glance)
Evidence
+1 0 −1 stress 30 min → Pearson r · P1·P2 ≈10–15 min arsenite osmotic
arsenite osmotic
illustrative · unpublished — workshop prototype

Across both stresses, the correlation between the two granule proteins climbs from negative to positive over ~10–15 min — they first exclude each other, then co-localize as the granule matures.

  • HeLa · sodium-arsenite and osmotic stress
  • Can't yet tell two merging droplets from one sub-compartmented droplet — needs a follow-up
  • The two stresses differ significantly in the first 10–15 min
S Sean · Vogel Lab · shared with Kate's lab · updated Oct 14
Method & context
Open to see context table, segmentation threshold, and pointer chips.
Open full study details →
Method is one click away — the <details> is native; works with JS disabled.
Method open (reproducer)
Evidence
+1 0 −1 stress 30 min → Pearson r · P1·P2 ≈10–15 min arsenite osmotic
arsenite osmotic
illustrative · unpublished — workshop prototype

Across both stresses, the correlation between the two granule proteins climbs from negative to positive over ~10–15 min — they first exclude each other, then co-localize as the granule matures.

  • HeLa · sodium-arsenite and osmotic stress
  • Can't yet tell two merging droplets from one sub-compartmented droplet — needs a follow-up
  • The two stresses differ significantly in the first 10–15 min
S Sean · Vogel Lab · node sg-composition-019ea6d6
Method & context

Question When stress granules form, do the two protein components condense together, or separately and then mix?

ParameterValue
Cell lineHeLa
Stresssodium arsenite · osmotic
ChannelsProtein 1–GFP (488) · Protein 2–mKate (561)
ImagingLattice SIM · ZEISS Elyra 7
Readoutper-granule co-localization vs time
Segmentation on per-pixel SUM · load-bearing
The segmentation choice is load-bearing. Segmenting on the per-pixel SUM of both channels (not one), thresholded just above background, defines every granule — and so every co-localization number that follows.
gitvogel-lab/sg-imaging · segment_stress_granules.ijm datagranule_composition_by_time.csv rawNDTiff z-stacks ≈ 80 GB · Elyra 7 video2-min walkthrough
Open full study details →
The threshold row pulses once to draw the reproducer's eye. Context table embedded in the card — no hop to a separate Protocol node.
Candidate (pre-evidence)
Candidate pre-formalized · Vogel Lab

Early sub-10-min granules — couldn't resolve two touching droplets from one at this z-spacing.

  • Co-localization too noisy below ~5 min — too few granules
  • 300 nm z-steps may merge the smallest adjacent granules
S Sean · Vogel Lab · architecture test in progress
Troubleshooting: trying a finer z-step and the FRAP readout before calling it — the SUM-segmentation may be merging adjacent granules.
Full troubleshooting history is preserved under the promoted node. The Convert button changes the card's type live (try it).
With discussion
Evidence
+1 0 −1 stress 30 min → ≈10–15 min arsenite osmotic
arsenite osmotic
illustrative · unpublished

Across both stresses, the correlation between the two granule proteins climbs from negative to positive over ~10–15 min — they first exclude each other, then co-localize.

  • HeLa · sodium-arsenite and osmotic stress
  • The segmentation choice (SUM of both channels) is load-bearing
S Sean · Vogel Lab · updated Oct 14
Method & context
Expand to see context table, threshold, and pointer chips.
Open full study details →
💬 Discussion (3)
K
Kate · 2d
Why segment on the SUM of the two channels, and not each one separately?
S
Sean · 1d
Per-channel biases toward whichever protein is brighter — the SUM keeps a granule a granule even while one component is still being excluded.
K
Kate · 4h
Makes sense. Can you push the time-course below 5 min? The early gap between the two stresses is the interesting part.
Viewing only — sign in to comment
Thread stays separate from the result and method body — nothing gets buried. Try "Convert to request" on Kate's last comment.
Evidence card · 4 states · rendered from MIRA JSON-LD over KOI · static-first, JS for State 3 & 4 interactions
read-only viewing surface