Skip to main content
. 2020 Jan 16;16(1):e9111. doi: 10.15252/msb.20199111

Figure 4. Proteome analysis of tumor growth patterns in pulmonary adenocarcinoma (ADC) FFPE tissue.

Figure 4

  1. Schematic illustration of the sample collection. Samples were collected from eight different patient tumors. For each tumor, sections were processed with hematoxylin and eosin (H&E) staining to locate different growth patterns of lepidic (low grade; group 1), acinar and papillary (intermediate grade; group 3), and solid and micropapillary (high grade; group 3). Two to four growth patterns per tumor were selected and sectioned into two consecutive 5 μm iterations, to provide replicates with highest possible similarity, resulting in a total of 51 samples (one iteration was missing).
  2. t‐Distributed stochastic neighbor embedding (t‐SNE) analysis of the proteome data corrected via a linear regression model. The different growth patterns are color‐coded as in panel A.
  3. Volcano plot showing differential expression analysis using Limma‐moderated t‐statistics for the comparison of lepidic samples against all other samples. Proteins passing significance thresholds of −log10 P < 0.05 (Benjamini–Hochberg adjusted) and an absolute log2 fold change of > 1 are highlighted in orange.
  4. Summary of significantly expressed proteins in the comparison of each growth indicated pattern against all others.

Source data are available online for this figure.