Accession number BBBC026 · Version 1
This 384-well plate has images of co-cultured hepatocytes and fibroblasts. Every other well is populated (A01, A03, ..., C01, C03, ...) such that 96 wells comprise the data. Each well has 9 sites and thus 9 images associated, totaling 864 images.
For each well there is one field and a single image nuclear image (Hoecsht). Images are in 8-bit PNG format.
CellProfiler pipeline before Cellprofiler Analyst ('Previous Workflow')
CellProfiler pipeline after ilastik ('New Workflow')
The images in column '01' have 4000 hepatocytes per well. These can be considered as 'negative' controls. The wells are specified in the filename.
The images in column '23' have 9500 hepatocytes per well. These can be considered as 'positive' controls. The wells are specified in the filename.
Hand-annotated Ground Truth Images
As described in section 2.4 of the paper noted below (Logan), 5 images were manually annotated with cell type. For each nucleus a colored seed point is given. Red seeds denote hepatocyte cells, green seeds denote fibroblast cells, and blue seeds denote other cell types.
These images are used in the CellProfiler pipelines comparing the performance of the two different workflows.
To replicate Figure 4 (in the article below)
Run these Matlab scripts
Further instructions on how to configure are in the comments.
Published results using this image set
Logan DJ, Shan J, Bhatia SN, Carpenter AE (2015). Quantifying co-cultured cell phenotypes in high-throughput using pixel-based classification. Methods pii: S1046-2023(15)30170-5 / doi. PMID: 26687239. PMCID: PMC4766037. (Research article)