MattDeVries commited on
Commit
3393ee1
·
verified ·
1 Parent(s): 374abac

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +33 -0
README.md ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ {}
3
+ ---
4
+
5
+ # ATLAS-1: 3D Drug-Treated Cancer Cells
6
+
7
+ **ATLAS-1** is a research dataset of single-cell **3D morphologies** captured under different drug treatments.
8
+ It contains **1,500** WM266-4 human melanoma cells embedded in **collagen** and imaged using **oblique-plane light-sheet microscopy**. Each cell is released as a **1024-point point cloud** sampled from a reconstructed **surface mesh**.
9
+
10
+ - **Total cells:** 1,500
11
+ - **Classes / conditions:**
12
+ - `nocodazole` (500) — microtubule inhibitor; rounded morphology
13
+ - `blebbistatin` (500) — myosin II inhibitor; elongated/spindle morphology
14
+ - `control` (500) — untreated baseline variability
15
+ - **Representations:** point clouds (1024 pts/cell) and watertight meshes
16
+ - **Intended use:** benchmarking point-cloud models on real biological shapes; studying drug-induced 3D morphology
17
+
18
+ > This dataset accompanies the **PointMIL** project (inherently interpretable point-cloud classification via MIL).
19
+
20
+ ---
21
+
22
+ ## Dataset Summary
23
+
24
+ Cells were segmented, meshed with **marching cubes** (+ Laplacian smoothing), then **uniformly sampled** to 1024 points per cell from the surface mesh. Files are organized by condition with a metadata CSV that holds labels and basic experiment info.
25
+
26
+ **Why ATLAS-1?**
27
+ - Real, noisy, heterogeneous 3D cell shapes (not synthetic CAD)
28
+ - Clear treatment effects (rounded vs elongated) for interpretable benchmarks
29
+ - Comes with both **mesh** and **point-cloud** views for flexible pipelines
30
+
31
+ ---
32
+
33
+ ## Repository Layout