t-distributed stochastic neighbor embedding (t-sne) spatial segmentation  (MathWorks Inc)

Journal: Journal of Pharmaceutical Analysis

Article Title: Discovering metabolic vulnerability using spatially resolved metabolomics for antitumor small molecule-drug conjugates development as a precise cancer therapy strategy

doi: 10.1016/j.jpha.2023.02.010

Figure Lengend Snippet: Heterogeneity of tumor histomorphology and tumor metabolic phenotype. (A) Haematoxylin and eosin (H&E) stained image of Lewis lung cancer. (B) Left: Image of tumor tissue spatial segmentation through automatic pixel labeling using t -distributed stochastic neighbor embedding ( t -SNE). Right: t -SNE 3D plot visualizing the distribution of four clusters. (C) Mass spectra of the four microregions in A. (D) Mass spectrometry images of representive metabolites in four types of cancers. The area drawn by the red and blue line refers to the cancerous area and paracancerous area, respectively; TCA: tricarboxylic acid; CHK: choline kinase.

Article Snippet: The t -distributed stochastic neighbor embedding ( t -SNE) spatial segmentation was performed using MATLAB 2018a (MathWorks, Natick, MA, USA) with self-written code ( ).

Techniques: Staining, Labeling, Mass Spectrometry