Using Visualization of <i>t</i>-Distributed Stochastic Neighbor Embedding To Identify Immune Cell Subsets in Mouse Tumors

Acuff, Nicole; Linden, Joel

doi:10.4049/jimmunol.1602077

Cited by 25 publications

(21 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the complexity of representing multiple immune subsets comprehensively, we utilized t stochastic neighborhood embedding (tSNE) to map high dimensional data onto two dimensional graphs [38]. tSNE analysis visually demonstrates strong myeloid subtype variation correlating to treatment (Fig.…”

Section: Resultsmentioning

confidence: 99%

Tumor microenvironment modulation enhances immunologic benefit of chemoradiotherapy

Hanoteau

Newton

Krupar

et al. 2019

j. immunotherapy cancer

View full text Add to dashboard Cite

BackgroundChemoradiotherapy (CRT) remains one of the most common cancer treatment modalities, and recent data suggest that CRT is maximally effective when there is generation of an anti-tumoral immune response. However, CRT has also been shown to promote immunosuppressive mechanisms which must be blocked or reversed to maximize its immune stimulating effects.MethodsTherefore, using a preclinical model of human papillomavirus (HPV)-associated head and neck squamous cell carcinoma (HNSCC), we developed a clinically relevant therapy combining CRT and two existing immunomodulatory drugs: cyclophosphamide (CTX) and the small molecule inducible nitric oxide synthase (iNOS) inhibitor L-n6-(1-iminoethyl)-lysine (L-NIL). In this model, we treated the syngeneic HPV-HNSCC mEER tumor-bearing mice with fractionated (10 fractions of 3 Gy) tumor-directed radiation and weekly cisplatin administration. We compared the immune responses induced by CRT and those induced by combinatory treatment (CRT + CTX/L-NIL) with flow cytometry, quantitative multiplex immunofluorescence and by profiling immune-related gene expression changes.ResultsWe show that combination treatment favorably remodels the tumor myeloid immune microenvironment including an increase in anti-tumor immune cell types (inflammatory monocytes and M1-like macrophages) and a decrease in immunosuppressive granulocytic myeloid-derived suppressor cells (MDSCs). Intratumoral T cell infiltration and tumor antigen specificity of T cells were also improved, including a 31.8-fold increase in the CD8+ T cell/ regulatory T cell ratio and a significant increase in tumor antigen-specific CD8+ T cells compared to CRT alone. CTX/LNIL immunomodulation was also shown to significantly improve CRT efficacy, leading to rejection of 21% established tumors in a CD8-dependent manner.ConclusionsOverall, these data show that modulation of the tumor immune microenvironment with CTX/L-NIL enhances susceptibility of treatment-refractory tumors to CRT. The combination of tumor immune microenvironment modulation with CRT constitutes a translationally relevant approach to enhance CRT efficacy through enhanced immune activation.Electronic supplementary materialThe online version of this article (10.1186/s40425-018-0485-9) contains supplementary material, which is available to authorized users.

show abstract

Section: Resultsmentioning

confidence: 99%

Tumor microenvironment modulation enhances immunologic benefit of chemoradiotherapy

Hanoteau

Newton

Krupar

et al. 2019

j. immunotherapy cancer

View full text Add to dashboard Cite

show abstract

“…Mixed model and Friedman test were used to investigate the time effect within the group when applicable. Heat map of normalized fold change versus the healthy controls of median cytokine concentration in T helper subsets and volcano plots were generated using R. Composition of CD4 T cells was analyzed using t-distributed stochastic neighbor embedding (t-SNE) 33,34 and automatic clustering by Matlab to visualize high-dimensional data. Composition of Tregs was analyzed using Uniform Manifold Approximation and Projection (UMAP) (https://doi.org/10.1101/ 298430) 69 and automatic clustering by phenograph to visualize high-dimensional data.…”

Section: Discussionmentioning

confidence: 99%

“…It was thus investigated whether the source of Th2 cytokines were adaptive Th2 cells or innate ILC2. The composition of CD4 T cells during asthma was generated using t-SNE 33,34 which utilizes unbiased clustering to visualize high-dimensional data. The identification of regulatory and helper T cells enabled us to visualize classical and non-classical subsets during acute and stable asthma (Fig.…”

Section: Modulation Of Helper Cd4 T Cells During Asthmamentioning

confidence: 99%

Dynamics of helper CD4 T cells during acute and stable allergic asthma

Kared

Tan

et al. 2018

Mucosal Immunology

View full text Add to dashboard Cite

Asthma comprises heterogeneous clinical subtypes driven by diverse pathophysiological mechanisms. We characterized the modulation of the inflammatory environment with the phenotype, gene expression, and function of helper CD4 T cells among acutely exacerbated and stable asthma patients. Systemic Th2 immune deviation (IgE and Th2 cytokines) and inflammation (IL-6, CRP) were associated with increased Th17 cells during acute asthma. Th2/Th17 cell differentiation during acute asthma was regulated by the enhanced expression of transcription factors (c-MAF, IRF-4). The development of pathogenic Th2 cells during acute asthma was characterized by the secretion of inflammatory cytokines coupled with Th2 molecules and PPARγ expression. The acquisition of CD15S, CD39, CD101, and CCR4 contributed to the increased heterogeneity of Regulatory T cells during asthma. Two clusters were derived from above cytokines, CD4 T cell phenotypes, and clinical data. Cluster 1, characterized by high eosinophils, Th2 and ILC2 frequencies, and higher exacerbation rates, may represent Th2-high subtype. Cluster 2 represents a more complex subtype; it is constituted by higher neutrophils or Th17 frequencies, higher inhaled corticosteroids dose and poor asthma control. In conclusion, we characterized systematically and longitudinally Th2-high and non-Th2 asthma subtypes and the heterogeneity of CD4 T cells in stable and acute asthma.

show abstract

“…https://doi.org/10.1371/journal.pone.0242301.t006 We visualized the deepest convolutional layer feature embedding for the ResNet-18 finetuned model, using the t-Distributed Stochastic Neighbor Embedding (t-SNE) algorithm [49], which is shown in Section D of the S1 File. The performance obtained with the fine-tuned models is compared to the Baseline, as shown in Table 6.…”

Section: Plos Onementioning

confidence: 99%

Analyzing inter-reader variability affecting deep ensemble learning for COVID-19 detection in chest radiographs

et al. 2020

View full text Add to dashboard Cite

Data-driven deep learning (DL) methods using convolutional neural networks (CNNs) demonstrate promising performance in natural image computer vision tasks. However, their use in medical computer vision tasks faces several limitations, viz., (i) adapting to visual characteristics that are unlike natural images; (ii) modeling random noise during training due to stochastic optimization and backpropagation-based learning strategy; (iii) challenges in explaining DL black-box behavior to support clinical decision-making; and (iv) inter-reader variability in the ground truth (GT) annotations affecting learning and evaluation. This study proposes a systematic approach to address these limitations through application to the pandemic-caused need for Coronavirus disease 2019 (COVID-19) detection using chest X-rays (CXRs). Specifically, our contribution highlights significant benefits obtained through (i) pretraining specific to CXRs in transferring and fine-tuning the learned knowledge toward improving COVID-19 detection performance; (ii) using ensembles of the fine-tuned models to further improve performance over individual constituent models; (iii) performing statistical analyses at various learning stages for validating results; (iv) interpreting learned individual and ensemble model behavior through class-selective relevance mapping (CRM)-based region of interest (ROI) localization; and, (v) analyzing inter-reader variability and ensemble localization performance using Simultaneous Truth and Performance Level Estimation (STAPLE) methods. We find that ensemble approaches markedly improved classification and localization performance, and that inter-reader variability and performance level assessment helps guide algorithm design and parameter optimization. To the best of our knowledge, this is the first study to construct ensembles, perform ensemble-based disease ROI localization, and analyze inter-reader variability and algorithm performance for COVID-19 detection in CXRs.

show abstract

Using Visualization of t-Distributed Stochastic Neighbor Embedding To Identify Immune Cell Subsets in Mouse Tumors

Cited by 25 publications

References 39 publications

Tumor microenvironment modulation enhances immunologic benefit of chemoradiotherapy

Tumor microenvironment modulation enhances immunologic benefit of chemoradiotherapy

Dynamics of helper CD4 T cells during acute and stable allergic asthma

Analyzing inter-reader variability affecting deep ensemble learning for COVID-19 detection in chest radiographs

Contact Info

Product

Resources

About