Dataset

BBBC013 dataset: U2OS cytoplasm to nucleus translocation (FKHR-GFP)

BBBC013 from the Broad Bioimage Benchmark Collection: U2OS cells expressing FKHR-EGFP, treated with Wortmannin or LY294002 dose curves on a 96-well plate, 640 x 640 images, with platemaps. CC BY 3.0. Browse all 192 images in DataTorch.

Contents

  1. What it is
  2. Files
  3. Browse it in DataTorch
  4. What people use it for
  5. Source and citation
Contents

What it is

A BBBC013 positive-control well (D12, 150 nM Wortmannin) open in DataTorch: FKHR-GFP gathered in the nuclei, with the drug and dose in the File Metadata panel
A BBBC013 GFP image, from the hosted copy

A translocation assay: when the drug acts, FKHR-EGFP moves from the cytoplasm into the nucleus. Imaged on an IN Cell Analyzer 3000 with its Trafficking Data Analysis Module.

FactValue
AccessionBBBC013, version 1
Cellshuman U2OS osteosarcoma, stably transfected with FKHR-EGFP
Channels1: FKHR-GFP, 2: DNA (DRAQ)
Plate96 wells, 9 two-fold dilutions (columns 3 to 11) per drug, 4 replicates of each curve
DrugsWortmannin (rows A to D), LY294002 (rows E to H)
Format640 x 640 px, FRM (77 MB) or 8-bit BMP (31 MB)
Images192 in the BMP set (96 wells x 2 channels), counted in the hosted copy
LicenseCC BY 3.0 (Ilya Ravkin)

Files

  • BBBC013_v1_images_frm.zip, BBBC013_v1_images_bmp.zip
  • BBBC013_v1_platemap_all.txt, BBBC013_v1_platemap_wortmannin.txt, BBBC013_v1_platemap_ly294002.txt
  • BBBC013_reproduce_logan.zip (5.4 MB), CellProfiler pipelines and metadata to reproduce the published analysis

Browse it in DataTorch

Dataset · datatorch/translocation-bbbc013 · 192 files

U2OS FKHR-GFP translocation (BBBC013)

Cytoplasm-to-nucleus translocation assay from the Broad Bioimage Benchmark Collection (BBBC013v1). 192 fluorescence microscopy images (640 x 640, 8-bit BMP) from one 96-well plate of human U2OS cells: 96 wells x 2 channels (channel 1 FKHR-GFP, channel 2 DNA stained with DRAQ), one field per well. The plate holds four replicate 9-point dose series of the PI3K inhibitors wortmannin (rows A-D, nM) and LY294002 (rows E-H, uM) plus 8 negative-control and 8 positive-control wells (150 nM wortmannin). Ground truth is the platemap: well, drug, dose and control role are stored as file metadata on every image and the control role is also an image-level label (negative-control, positive-control, dose). The two channels of each well are separate files paired through metadata and folder path; there are no cell outlines. License: CC BY 3.0 (Ilya Ravkin). Source: https://bbbc.broadinstitute.org/BBBC013. Please cite: We used image set BBBC013v1 provided by Ilya Ravkin, available from the Broad Bioimage Benchmark Collection [Ljosa et al., Nature Methods, 2012].

Open →

The hosted copy holds the whole 8-bit BMP set: 192 images, 96 wells in two channels, one field per well. The platemap travels with the images, so you never have to join a text file by hand:

  • Drug, dose and well are stored as metadata on every image. Filter by drug or dose to pull one curve, or one concentration across the replicates.
  • Control role is also an image-level label with three values: negative-control (16 images, 8 wells), positive-control (16 images, 8 wells) and dose (160 images, 80 wells).
  • The two channels of a well are separate files, paired through their metadata and folder path. There are no cell outlines, so this is a dose-response and classification set, not a segmentation one.

One difference from the platemap file: for wells E01 to H01 the hosted copy follows the metadata in BBBC013_reproduce_logan.zip, which marks them as Wortmannin controls. The platemap file's original value for those four wells is kept in a platemap_dose field, so both readings are there.

What people use it for

Checking that a translocation measure (nuclear over cytoplasmic GFP) gives a clean dose response, and comparing assay quality across methods.

Source and citation

Source: https://bbbc.broadinstitute.org/BBBC013 (read 2026-10-22, license re-checked 2026-10-29). Cite: "We used image set BBBC013v1 provided by Ilya Ravkin, available from the Broad Bioimage Benchmark Collection [Ljosa et al., Nature Methods, 2012]."

Annotate your own screening images

Review multi-channel plates and label cells with your team in the browser.

Frequently asked questions

What is the ground truth in BBBC013?

The dose of each well. Platemap files in CellProfiler LoadText format list every sample, the Wortmannin wells (rows A to D) and the LY294002 wells (rows E to H). Negative controls have no drug. Positive controls have 150 nM Wortmannin.

What image formats does BBBC013 come in?

Native FRM (77 MB zip) or 8-bit BMP (31 MB zip), 640 x 640 px, one image per channel.

How many images are in BBBC013?

The BMP set has 192 images, one field per well for 96 wells, in two channels (FKHR-GFP and DNA). That is the count of the hosted copy, which holds the whole BMP set.

What is the license of BBBC013?

Creative Commons Attribution 3.0 Unported (CC BY 3.0), by Ilya Ravkin.

Get the data

Open the hosted copy on DataTorch.

Browse the files in your browser, then download the dataset with a free account.

Open the datasetStart free

Related

Tools and community for image annotation: label, review, and share datasets, from everyday photos to the imagery only specialists can read.

Platform

PlatformImaging labsExplore datasetsEnterprisePricing

© 2026 DataTorch. All rights reserved.