![Figure 2. GSEA ES curves and clustering results for the 2 significant gene signatures significantly overexpressed in AITL compared to PTCL-u. (A-B) ES curve (red) or the running sum of the weighted enrichment score obtained from GSEA software. Vertical blue lines indicate the position of each of the 100 genes (“Hits”) comprising the GS2_ TFH _up gene set. The graph on the bottom of each panel shows the ranked list metric (gray, SNR for A and fold change for B) for each gene as a function of the rank in the ordered data set. This corresponds to the level of correlation with the groups being tested (the far leftmost gene had the highest correlation with the AITL samples, and the far rightmost gene had the highest correlation with PTCL-u samples). (A) Results obtained for GS2_ TFH _up gene set in the comparison of AITL and PTCL-u using the SNR statistic to rank the genes from left to right (highest [0.9] to the lowest [−0.4] real SNR value, respectively). (B) Results obtained for the 42 core (or leading-edge) genes obtained from the analysis in panel A that were distributed in the top 1722 genes. The core genes in the ranked list represent the genes that are at, or before, the point where the running sum reaches its maximum deviation from zero (see Subramanian et al27 for more details). Here, the top 1722 genes obtained in panel A are ranked from highest (left) to lowest (right) fold change between the average of the 2 AITL-sorted cell samples (AITL cell) versus the average of the 17 AITL tissue samples and the hits represent the 42 core genes. (C) Clustering and heatmap of these 42 core genes obtained from the analysis shown in panel A. Data from T-cell subsets (TFH, Th1, and Th2; Chtanova et al21 and this study) were standardized by calculating an independent z-score (mean = 0 and standard deviation = 1) for each data set. Each gene was further standardized by mean-centering its expression profile across all samples. Genes are ordered by their rank calculated in the analysis shown in panel A. DNA-Chip Analyzer (dChip) Version 1.3 (Department of Biostatistics, Harvard School of Public Health, Boston, MA) was used to generate the heatmap and cluster the genes. Standardized expression ranges from −2.0 (blue) to 2.0 (red). Arrows point at the columns representing AITL cell suspensions.](https://ash.silverchair-cdn.com/ash/content_public/journal/blood/109/11/10.1182_blood-2006-10-055145/4/zh80110703680002.jpeg?Expires=1723224131&Signature=HGRD0-MDMG4LTSYY0Fwg5m7dlJAfQE9iesEgFsvOIIqqWoyVKtUQHPTAs3GGH7exo~DPFg9bDdsLMySomf4~Wj~VwAjow1GTR9WBpZxaSJjBRmyW2sH508FGF2Z058oiCBv1ngcYUx-BVKx~183xxxTTtwClBH~Vuta1bsRzAZvdCtZej42XLwOgnWqeJnnxz0IgE5MA9CXaT7peqCI9Gze4jpidiHdQbRd5~JOE9XiWvisb2I94rinOQNi0Q1-lxj7~~3rIEwCXp3j4CtesUNbux~XKAeq7JHQnXpN-ViJkggb7-Z-7RPBmIKBihWzQe2KzJfKtC8oTyKgwvbhnuA__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
GSEA ES curves and clustering results for the 2 significant gene signatures significantly overexpressed in AITL compared to PTCL-u. (A-B) ES curve (red) or the running sum of the weighted enrichment score obtained from GSEA software. Vertical blue lines indicate the position of each of the 100 genes (“Hits”) comprising the GS2_ TFH _up gene set. The graph on the bottom of each panel shows the ranked list metric (gray, SNR for A and fold change for B) for each gene as a function of the rank in the ordered data set. This corresponds to the level of correlation with the groups being tested (the far leftmost gene had the highest correlation with the AITL samples, and the far rightmost gene had the highest correlation with PTCL-u samples). (A) Results obtained for GS2_ TFH _up gene set in the comparison of AITL and PTCL-u using the SNR statistic to rank the genes from left to right (highest [0.9] to the lowest [−0.4] real SNR value, respectively). (B) Results obtained for the 42 core (or leading-edge) genes obtained from the analysis in panel A that were distributed in the top 1722 genes. The core genes in the ranked list represent the genes that are at, or before, the point where the running sum reaches its maximum deviation from zero (see Subramanian et al27 for more details). Here, the top 1722 genes obtained in panel A are ranked from highest (left) to lowest (right) fold change between the average of the 2 AITL-sorted cell samples (AITL cell) versus the average of the 17 AITL tissue samples and the hits represent the 42 core genes. (C) Clustering and heatmap of these 42 core genes obtained from the analysis shown in panel A. Data from T-cell subsets (TFH, Th1, and Th2; Chtanova et al21 and this study) were standardized by calculating an independent z-score (mean = 0 and standard deviation = 1) for each data set. Each gene was further standardized by mean-centering its expression profile across all samples. Genes are ordered by their rank calculated in the analysis shown in panel A. DNA-Chip Analyzer (dChip) Version 1.3 (Department of Biostatistics, Harvard School of Public Health, Boston, MA) was used to generate the heatmap and cluster the genes. Standardized expression ranges from −2.0 (blue) to 2.0 (red). Arrows point at the columns representing AITL cell suspensions.
