Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2021 Feb 22;49(10):e55. doi: 10.1093/nar/gkab095

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© The Author(s) 2021. Published by Oxford University Press on behalf of Nucleic Acids Research.

This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com

PMC Copyright notice

Figure 1. — Discovery of image-integrated spatially variable genes and functional terms in breast cancer data. (A) Multiple patches were extracted from a tissue slide image based on the coordinates of sampling spots. Each image patch was provided as an input to the pretrained convolutional neural network (CNN) model, VGG-16. A total of 512 image features extracted from the CNN were further processed with principal component analysis (PCA) to reduce the dimensions. SPADE genes were constructed by a linear model to identify gene expression correlated with the PC image latent of each spot. (B) Spatial mapping of the PC1, PC2 and PC3 image latent from breast cancer tissue. The PC values of each spot are visualized using colormaps. The maximum and minimum values of the colormap represent two standard deviations above and below the mean value, respectively. (C) Volcano plots for highly associated genes with PC1 image latent features. The cutoff for the log₂ regression coefficient (RC) and adjusted P-value (Benjamini–Hochberg correction) is 0.55 and 10⁻¹⁵_, respectively. Spatial expression of the top four genes representing the greatest contrast in the (D) PC1 and (E) PC3 image latent space. The top genes are presented in descending order of |log₂RC| (FDR < 0.05). The normalized gene expression level of each spot is visualized with colormaps. The maximum and minimum values of the colormap represent two standard deviations above and below the mean expression, respectively. Gene ontology (GO) analysis for (F) PC1 and (G) PC3 SPADE genes showing positive or negative association with PC image latent in breast cancer data. The top 3 positive or negative GO terms for each subcategory, biological process (BP), cellular component (CC) and molecular function (MF), are exhibited in the left and right panel, respectively. The number of overlapping genes is expressed as the size of the dot, and the Benjamini-Hochberg adjusted P-value is exhibited with a colormap.