Nucleus feature profiles from GTEx histological images across 12 tissues
Data files
Sep 29, 2025 version files 217.41 GB
-
DE_genes.zip
49.90 MB
-
EWAS.zip
20.03 MB
-
qupath-artery-tibial.tar.gz
5.82 GB
-
qupath-esophagus-mucosa.tar.gz
19.86 GB
-
qupath-esophagus-muscularis.tar.gz
26.16 GB
-
qupath-heart-atrial-appendages.tar.gz
8.30 GB
-
qupath-heart-leftventricle.tar.gz
12.68 GB
-
qupath-lung.tar.gz
31.86 GB
-
qupath-muscle.tar.gz
14.19 GB
-
qupath-nerve-tibial.tar.gz
5.31 GB
-
qupath-skin-not-exposed.tar.gz
13.16 GB
-
qupath-skin-sun-exposed.tar.gz
14.03 GB
-
qupath-testis.tar.gz
24.71 GB
-
qupath-thyroid.tar.gz
41.25 GB
-
README.md
6.94 KB
-
Summary_statistics_of_QuPath-extracted_nucleus_features.xlsx
2.88 MB
Abstract
This dataset contains nucleus features extracted from GTEx histological images using QuPath across 12 human tissues. It also includes summary statistics of these features, which were applied in image quantitative trait loci (imageQTL) analyses and differential expression (DE) analyses. The dataset enables the exploration of relationships between nuclear morphology, genetic variation, and gene expression. By providing feature measurements across multiple tissues, it serves as a resource for integrative studies in computational pathology, imaging genomics, and functional genomics. The histological image data used in these analyses were obtained from the GTEx Portal on 10/01/2022, and we acknowledge the GTEx Consortium for providing these publicly available resources.
Dataset DOI: 10.5061/dryad.8gtht771x
Description of the data and file structure
This dataset contains nucleus features extracted from GTEx whole-slide images (WSIs) using QuPath v0.4.3, enabling quantitative analyses of tissue morphology across multiple human tissues.
The dataset includes a total of 15 files: 12 tissue files, each corresponding to a single tissue type and stored in a compressed folder, containing nucleus feature data for individual donors derived from histological image segmentation; one summary statistics file that provides aggregated statistics across all 12 tissues; one compressed file containing differentially expressed genes for various nucleus feature groups; and one file with epigenome-wide association results linking nucleus features to methylation loci.
Key references for the dataset include QuPath (https://github.com/qupath/qupath/releases) and its tutorial (https://qupath.readthedocs.io/en/stable/docs/tutorials/index.html), as well as the GTEx Portal (https://gtexportal.org/home/), where the original WSIs were obtained.
Files and variables
The 12 compressed folders, named qupath-<tissue-name>.tar.gz, containing image features extracted with QuPath for each tissue, including artery tibial, esophagus mucosa, esophagus muscularis, heart atrial appendages, heart left ventricle, lung, muscle, tibial nerve, skin (not sun-exposed), skin (sun-exposed), testis, and thyroid. Within each compressed folder, image features for each donor are stored as individual files, named <sample-name>_Detections.txt. In each donor file, rows correspond to detected cells and columns correspond to image features. The raw files contain various types of features extracted by QuPath, but only nucleus-related features were used in this study.
File: qupath-artery-tibial.tar.gz
Description: Nuclear features extracted from QuPath for tibial artery tissue samples from GTEx.
File: qupath-esophagus-mucosa.tar.gz
Description: Nuclear features extracted from QuPath for esophagus mucosa tissue samples from GTEx.
File: qupath-esophagus-muscularis.tar.gz
Description: Nuclear features extracted from QuPath for esophagus muscularis tissue samples from GTEx.
File: qupath-heart-atrial-appendages.tar.gz
Description: Nuclear features extracted from QuPath for heart atrial appendages tissue samples from GTEx.
File: qupath-heart-leftventricle.tar.gz
Description: Nuclear features extracted from QuPath for heart left ventricle tissue samples from GTEx.
File: qupath-lung.tar.gz
Description: Nuclear features extracted from QuPath for lung tissue samples from GTEx.
File: qupath-muscle.tar.gz
Description: Nuclear features extracted from QuPath for muscle tissue samples from GTEx.
File: qupath-nerve-tibial.tar.gz
Description: Nuclear features extracted from QuPath for tibial nerve tissue samples from GTEx.
File: qupath-skin-not-exposed.tar.gz
Description: Nuclear features extracted from QuPath for skin (not-sun-exposed) tissue samples from GTEx.
File: qupath-skin-sun-exposed.tar.gz
Description: Nuclear features extracted from QuPath for skin (sun-exposed) tissue samples from GTEx.
File: qupath-testis.tar.gz
Description: Nuclear features extracted from QuPath for testis tissue samples from GTEx.
File: qupath-thyroid.tar.gz
Description: Nuclear features extracted from QuPath for thyroid tissue samples from GTEx.
File: Summary_statistics_of_QuPath-extracted_nucleus_features.xlsx
Description: Summary statistics of QuPath-extracted nucleus features across samples by tissue. The summary statistics Excel file contains 12 sheets, one for each tissue. Within each sheet, rows represent sample names, and 24 columns provide summary statistics — including mode, standard deviation, Q1, Q2, Q3, and mean — for four nucleus features (area, circularity, eccentricity, and nucleus-to-cell area ratio).
File: DE_genes.zip
Description: This zip file contains 48 CSV files with differential expression (DE) results for nucleus feature–based groups. The files are named as 'tissue'**'nucleus feature type'table.csv. DE genes were identified for each of the 12 GTEx tissues and for each of the four nucleus features (area, circularity, eccentricity, and nucleus-to-cell area ratio). For example, Thyroid_area_table.csv contains DE results for Thyroid, comparing samples with different nucleus sizes. For each whole-slide image (WSI), the mode value of each nucleus feature was used as a representative metric. GTEx individuals were then categorized into non-overlapping binary groups (‘large’ vs. ‘small’) based on these mode values within each tissue and feature type. For instance, in thyroid tissue, WSIs were classified into ‘large’ or ‘small’ nucleus area groups, and DE genes were calculated between these groups. The same approach was applied across all tissues and nucleus features.
File: EWAS.zip
Description: Epigenome-wide association analyses were performed using CHAMP to link nucleus features with DNA methylation loci. Six result files are provided, including DMP (Differentially Methylated Positions), DMR (differentially methylated regions), and GSEA (Gene Set Enrichment Analysis) outputs for lung nucleus eccentricity and nucleus-to-cell area ratio. The CSVs show raw genomic regions of altered methylation linked to nuclear features and the JSON files show biological interpretation, showing which pathways and gene networks are enriched among those regions.
Code/software
N/A
Access information
Other publicly accessible locations of the data:
Data was derived from the following sources:
Human subjects data
All data used in this study are derived from GTEx histological images that are publicly available on the GTEx website. These images were collected from GTEx donors under informed consent for research use. The images are fully de-identified by GTEx prior to public release, removing all personal identifiers and any information that could reasonably be used to trace back to individual donors. Our dataset contains only these de-identified, publicly accessible images and derived nucleus feature measurements, and therefore can be shared in the public domain without additional access restrictions.
QuPath v0.4.3 was used to extract nucleus features from GTEx WSIs. QuPath is publicly available (https://github.com/qupath/qupath/releases). Default settings were followed as outlined in the QuPath tutorial: https://qupath.readthedocs.io/en/stable/docs/tutorials/index.html. The image type was set to 'brightfield H&E,' and stain vectors were optimized. Subsequently, we ran a pixel classifier to identify relevant tissue areas. Cell and nucleus detection was performed using QuPath’s cell detection tool. Finally, the extracted cell and nucleus detection measurements for each WSI resulted in a structured dataset for further analysis. We configured the pipeline for each tissue type to accommodate different tissue morphologies. This automated feature extraction was crucial for ensuring consistent and objective quantification of tissue characteristics across all slides.
Four key nucleus features were included in the summary statistics: (1) nucleus area, (2) nucleus circularity, (3) nucleus eccentricity, and (4) nucleus-to-cell area ratio. We employed several parameters to characterize the distribution within each WSI. Outliers were filtered using a 1.5 interquartile range. We then applied kernel density estimation, partitioning the data into 500 bins, and selected the value corresponding to the highest density, representing the mode of the entire distribution. Additionally, we included the first quartile (Q1), second quartile (Q2), third quartile (Q3), mean (here we used a 90% trimmed mean), and standard deviation to describe the distribution. These six representative values for each nucleus feature were then utilized in subsequent analyses, resulting in 24 nucleus features in total.