Data from: Short activation domains control chromatin association of transcription factors
Data files
Jan 31, 2025 version files 592.54 MB
-
dryad_dataset.zip
592.54 MB
-
README.md
2.23 KB
Abstract
Transcription factors regulate gene expression with DNA-binding domains (DBDs) and activation domains. It is generally assumed that DBDs are solely responsible for interacting with DNA and chromatin. Here, we used single-molecule tracking of transcription factors in living cells to nd that short activation domains can control the fraction of molecules bound to chromatin. Activation domains with high bound fractions also have longer residence times on chromatin. Mutations in activation domains that increase activity of a transcriptional reporter increase the fraction of molecules bound to chromatin. Reciprocally, mutations that decrease reporter activity decrease fraction bound. These eects were consistent across three activation domains and three DBD classes. Taken together, these results suggest that activation domains play a major role in tethering transcription factors to chromatin, challenging the traditional view that the DBD is solely responsible for binding chromatin.
README: Dataset in support of "Short activation domains control chromatin association of transcription factors"
https://doi.org/10.5061/dryad.41ns1rnqt
Description of the data and file structure
Pre-processed SMT data ("trajectories") are organized by Figure in the folder "SMT_trajectories". (We could not deposit raw SMT movies due to storage limitations.) Raw FRAP movies (35) are in the folder "FRAP_movies".
Processing steps and how data presented in Figures can be generated are described in detail in the Methods section of the manuscript. Briefly, SMT movies are subjected to detection (which pixels have a spot?), subpixel localization (where is the spot center?), and tracking (which spots connect to which others in subsequent frames?). The "trajs.csv" files here describe localizations, one spot per line, and how they are connected. Data from many cells corresponding to one condition have been aggregated and are tabulated here as single CSVs. A Bayesian mixture model computes diffusion spectra based on these data.
For FRAP movies, custom python code is used to quantify spot recovery, accounting for spot drift, changes in nuclear intensity, and background intensity.
"trajs.csv" files
CSV files (within the folders named 2a, 3a, 3b, 3c, 4a, 4b, 4c, 5a, 5b) where each line describes a localization that has passed a spot detection threshold. Columns:
- y: sub-pixel y-coordinate of the spot
- x: sub-pixel x-coordinate of the spot
- I0: integrated intensity of the point-spread function fit to this spot
- bg: pixel-wise background of the point-spread function fit to this spot
- snr: maximum value of the log-likelihood test used to detect this spot
- y_detect: y-pixel where this spot was detected
- x_detect: x-pixel where this spot was detected
- frame: frame in which this spot was detected
- trajectory: index that assigns spots to trajectories. Spots with the same "trajectory" value are connected to each other.
- mask_index: index of the masked cell that gave rise to this spot. Not useful in the aggregated CSV.
CZI files
ZEISS file format for microscopy data. Contains metadata, can be read by open-source libraries, e.g. Python's czifile
.