Data from: Multiscale chromatin dynamics and high entropy in plant iPSC ancestors
Data files
Mar 26, 2024 version files 74.45 MB
-
HTI_001.csv
36.86 MB
-
HTI_002.csv
6.14 MB
-
HTI_004.csv
16.46 MB
-
HTI_005.csv
14.99 MB
-
README.md
3.08 KB
Abstract
Plant protoplasts provide a starting material to induce pluripotent cell masses in vitro competent for tissue regeneration. Dedifferentiation is associated with large-scale chromatin reorganisation and massive transcriptome reprogramming, characterized by stochastic gene expression. How this cellular variability reflects on chromatin organisation in individual cells and what are the factors influencing chromatin transitions during culturing is largely unknown. High-throughput imaging and a custom, supervised image analysis protocol extracting over 100 chromatin features unravelled a rapid, multiscale dynamics of chromatin patterns which trajectory strongly depends on nutrients availability. Decreased abundance in H1 (linker histones) is hallmark of chromatin transitions. We measured a high heterogeneity of chromatin patterns indicating an intrinsic entropy as hallmark of the initial cultures. We further measured an entropy decline over time, and an antagonistic influence by external and intrinsic factors, such as phytohormones and epigenetic modifiers, respectively. Collectively, our study benchmarks an approach to understand the variability and evolution of chromatin patterns underlying plant cell reprogramming in vitro.
README: Multiscale chromatin dynamics and high entropy in plant iPSC ancestors
https://doi.org/10.5061/dryad.pnvx0k6wp
The HTI datasets correspond to image features extracted from high-throughput imaging of Arabidopsis leaf protoplasts cultures (multi-well plate) expressing fluorescent chromatin markers and cultivated several days under distinct media. The experiments description of the culturing medium, specific treatment, the imaging day, replicate culture number and marker line is provided in Table S4 of the related paper and associated with this link.
The image features were extracted by TissueMaps (http://tissuemaps.org following segmentation of the cell's nuclei labelled with H2B-RFP (See the related paper). The features are derived from a package described by Hamilton et al (2007).
The data provide for each segmented nuclei: morphology descriptors of the segmented nuclei, signal intensity variables for each marker (H1.2-GFP and H2B-RFP), and texture features. There are four families of texture features (LBP, TAS, Gabor and Hu). Textures metrics describe the spatial distribution of signal intensities as a function of scale (Depeursinge et al., 2017a; Di Cataldo & Ficarra, 2017).
The images used for segmentation are available at [IDR or DRYAD link to the images, in preparation]
Description of the dataset structure
- Mapobject_id : unique identifier of segmented nucleus
- Day: Culturing day at imaging
- Treatment: compound added to the culturing medium (eg, TSA, phytohormones)
- Medium: culturing medium (Gamborg or W5, see Table S3 of related paper)
- Line: Plant line used to prepare the protoplast culture (see Table S4 of related paper)
- Well: well number on the multi-well plate used for plant cells cultures and imaging (see Methods in related paper)
- Columns with prefix Morphology_: morphological descriptors of the object (nucleus).
- Columns with prefix Intensity_: signal intensity descriptors (sum, mean, standard deviation) for each channel (C01= channel 1, H1.2-GFP ; C02= channel 2, H2B-RFP). In addition, Intensity_Mean_Ratio is the ratio of Intensity_Mean_C01/ Intensity_Mean_C02
- Columns with prefix Texture_: texture descriptors for 4 families of texture (LBP, Gabor, TAS and Hu, see Table S1 and main text of related paper)
Software
These data were exported from TissueMaps after segmentation.
http://docs.tissuemaps.org/latest/
References
Depeursinge, A., Fageot, J., & Al-Kadi, O. S. (2017b). Fundamentals of Texture Processing for Biomedical Image Analysis. 1-27. doi:10.1016/b978-0-12-812133-7.00001-6
Di Cataldo, S., & Ficarra, E. (2017). Mining textural knowledge in biological images: Applications, methods and trends. Comput Struct Biotechnol J, 15, 56-67. doi:10.1016/j.csbj.2016.11.002
Hamilton, N. A., Pantelic, R. S., Hanson, K., & Teasdale, R. D. (2007). Fast automated cell phenotype image classification. BMC Bioinformatics, 8, 110. doi:10.1186/1471-2105-8-110
Methods
Image features extracted by Tissue Maps from microscopy images following automated, supervised segmentation of plant protoplast nuclei.
Leaf protoplasts were cultured in coverglass-bottom 96-well plates and imaged using a confocal microscope Cell Voyager. The plant cell line (expressing specific nuclear reporters), imaging day (day0-day7) and culturing media is indicated in the dataset. Additional information is provided in the Table S3 in the manuscript.
Image features correspond to three families: intensity, morphology and texture. A complete list of image features used for the study is available as Table S2 in the manuscript.