t-SNE Analysis Research Articles (Page 10)

PURPOSEThis study aims to test whether MRI radiomic signatures can distinguish radiation necrosis (RN) from tumor progression (TP) in a multi-institution dataset using machine learning.METHODSBrain metastases treated with SRS were followed by serial MRI, and those showing evidence of RN or TP underwent pathologic confirmation. Radiomic features were extracted from T1 post-contrast (T1c) and T2 fluid attenuated inversion recovery (T2 FLAIR) MRI. High dimensional radiomic feature space was visualized in a two-dimensional space using t-distributed stochastic neighbor embedding (t-SNE). Cases from 2 institutions were combined and randomly assigned to training (2/3) and testing (1/3) cohorts. Backward elimination was used for feature selection, followed by random forest algorithm for predictive modeling.RESULTSA total of 135 individual lesions (37 RN and 98 TP) were included. The majority (72.6%) received single-fraction SRS to a median dose of 18Gy. Clear clustering of cases around the institutional origin was observed on t-SNE analysis. 21 T1c and 4 FLAIR features were excluded from subsequent modeling due to significant correlation with the institutional origin. Backward elimination yielded 6 T1c and 6 FLAIR features used for model construction. A random forest model based on the 6 FLAIR features (cluster shade, neighborhood gray tone difference matrix (NGTDM) coarseness, NGTDM texture strength, run length nonuniformity, run percentage, and short run high gray-level emphasis) achieved sensitivity of 76% and specificity of 70% on the training cohort (AUC 0.74, 95% CI 0.60–0.88), and sensitivity of 67% and specificity of 83% on the testing cohort (AUC 0.75, 95% CI 0.59–0.93). Addition of the T1c features resulted in overfitting of the training cohort (AUC 1.00), but did not improve model performance on the testing cohort (AUC 0.69, 95% CI 0.51–0.87).CONCLUSIONMRI radiomics based machine learning can distinguish RN from TP after brain SRS in a heterogeneous image dataset.

BackgroundSingle rare cell characterization represents a new scientific front in personalized therapy. Imaging mass cytometry (IMC) may be able to address all these questions by combining the power of MS-CyTOF and microscopy.MethodsWe have investigated this IMC method using < 100 to up to 1000 cells from human sarcoma tumor cell lines by incorporating bioinformatics-based t-Distributed Stochastic Neighbor Embedding (t-SNE) analysis of highly multiplexed IMC imaging data. We tested this process on osteosarcoma cell lines TC71, OHS as well as osteosarcoma patient-derived xenograft (PDX) cell lines M31, M36, and M60. We also validated our analysis using sarcoma patient-derived CTCs.ResultsWe successfully identified heterogeneity within individual tumor cell lines, the same PDX cells, and the CTCs from the same patient by detecting multiple protein targets and protein localization. Overall, these data reveal that our t-SNE-based approach can not only identify rare cells within the same cell line or cell population, but also discriminate amongst varied groups to detect similarities and differences.ConclusionsThis method helps us make greater inroads towards generating patient-specific CTC fingerprinting that could provide an accurate tumor status from a minimally-invasive liquid biopsy.

Mutations in KRAS, NRAS, and BRAF (RAS/BRAF) genes are the main predictive biomarkers for the response to anti-EGFR monoclonal antibodies (MAbs) targeted therapy in metastatic colorectal cancer (mCRC). This retrospective study aimed to report the mutational status prevalence of these genes, explore their possible associations with clinicopathological features, and build and validate a predictive model. To achieve these objectives, 500 mCRC Mexican patients were screened for clinically relevant mutations in RAS/BRAF genes. Fifty-two percent of these specimens harbored clinically relevant mutations in at least one screened gene. Among these, 86% had a mutation in KRAS, 7% in NRAS, 6% in BRAF, and 2% in both NRAS and BRAF. Only tumor location in the proximal colon exhibited a significant correlation with KRAS and BRAF mutational status (p-value = 0.0414 and 0.0065, respectively). Further t-SNE analyses were made to 191 specimens to reveal patterns among patients with clinical parameters and KRAS mutational status. Then, directed by the results from classical statistical tests and t-SNE analysis, neural network models utilized entity embeddings to learn patterns and build predictive models using a minimal number of trainable parameters. This study could be the first step in the prediction for RAS/BRAF mutational status from tumoral features and could lead the way to a more detailed and more diverse dataset that could benefit from machine learning methods.

3045 Background: Although anti-PD-1/PD-L1 therapy has become one of the standard treatments for advanced cancers, its low treatment efficacy (10-30%) has remained a major issue. We sought to perform a detailed immune profiling of cells and soluble proteins in order to characterize key regulators and signaling molecules and identify therapeutic targets and biomarkers that may improve treatment efficacy and diagnosis. Methods: This observational study enrolled 49 advanced cancer patients treated with PD-1/PD-L1 blockade monotherapy. Treatment response was assessed by RECIST 1.1. PBMC and plasma samples were collected at baseline and every 6 weeks following initial treatment. Immune profiling of PBMC was done by multi-parametric flow cytometer, and t-SNE analysis was used to identify key immune subtypes. Soluble proteins were evaluated by LUMINEX assays. Cut-off values were determined by ROC curve analysis. Results: Three unique subtypes of immune cells were identified. The population of CD11c+HLA-DRlowCD80+CD86− CD274+ cells (regDC) at baseline was significantly higher in patients with progressive disease (PD, n=28) than in patients showing clinical benefit (non-PD, n=21; p=0.030). The higher regDC population also correlated with higher levels of IL-8, IL-10, CXCL1, CXCL5, and CXCL11 in plasma. The population of CD4+CD25+CD62L+ T cells (Treg) was also higher in PD patients (p&lt;0.001). A unique subtype of CD4+CD28− T cell, however, was higher in non-PD patients (p&lt;0.001). For the soluble proteins, the levels of sLAG-3 and sGITR in plasma correlated with better clinical outcome in low regDC patients (p=0.004 and 0.044, respectively). The combined biomarker panel (cellular and protein markers) yields high sensitivity (90.5 %) and specificity (82.1 %) for predicting treatment efficacy. Disease control rate (DCR) and median progression free survival (PFS) are shown in the Table. Conclusions: To our knowledge, this pilot study is the first to detect three immune cell subtypes, regDC, Treg and CD4+CD28− cells, associated with clinical outcome in the treatment of PD-1/PD-L1 blockade. Profiling of immune cell subtypes and soluble immune checkpoint proteins can serve to identify therapeutic targets and biomarkers for treatment efficacy. We will report the data with further enrollment. [Table: see text]

3566 Background: Clinical biomarker studies are often hindered by the availability of tissue specimens of sufficient quality and quantity. While RNA-Seq is often considered the gold standard for measuring mRNA expression levels in cancer tissue, it typically requires multiple formalin-fixed paraffin-embedded (FFPE) tissue sections to extract a sufficient amount of quality RNA for subsequent gene expression profiling analysis. The HTG EdgeSeq technology is a gene expression profiling platform that combines quantitative nuclease protection assay technology with next-generation sequencing detection. Unlike RNA-Seq, the HTG EdgeSeq technology does not require RNA extraction, and can use small amounts of tissue material, typically several mm2, to generate reproducible gene expression profiles. Methods: This study compares the performance of RNA-Seq and HTG's profiling panel, the HTG EdgeSeq Precision Immuno-Oncology Panel (PIP), which is designed to measure expression levels of 1,392 genes focused on tumor/immune interaction. Approximately 1,200 samples from three tumor indications (gastric cancer, colorectal cancer and ovarian cancer) were tested using both technologies. Results: Up to four FFPE slides were used for RNA extraction to support RNA-Seq testing; out of the 1,202 samples processed, 1,099 generated extracted RNA of sufficient quality and quantity (as measured by RNA concentration, RIN score and %DV200) to proceed to sequencing, which resulted in a pass rate of 91.4% for RNA-Seq. The HTG EdgeSeq PIP panel resulted in a pass rate of 97.3% (samples passing QC metrics) when the same 1,200 samples were tested, and required only a single FFPE section owing to the small sample requirement. The t-SNE (a non-linear dimensionality reduction method) analysis of the common 1,358 genes revealed similar clustering of the three cancer indications between the two methods. Correlations across individual genes by sample resulted in the mean Spearman correlation coefficient of 0.73 (95% confidence interval of 0.61 - 0.80). Additionally, gene-wise comparisons across all samples were also evaluated. Conclusions: These data demonstrate that HTG EdgeSeq gene expression panels can be used as a competitive alternative to RNA-Seq, generating equivalent gene expression results, while offering the added benefits of a small sample size requirement, lack of RNA extraction bias, and fully automated data analysis pipeline.

5074 Background: Existing clinicopathological tools are unable to accurately identify renal cell carcinoma (RCC) patients who will develop metastases after surgery. As a result, it is unclear how long and how often to follow-up patients post-operatively. Tumor macropathology, as assayed by CT scanning, represents the sum product of tumor biology and microenvironment. We hypothesized that quantitative tumor features extracted from CT scans (termed radiomics) could discriminate between metastatic and non-metastatic RCCs. Methods: This retrospective study incorporated three cohorts of clear-cell RCC patients (n = 279, from TCGA, CPTAC and KiTS19 datasets) treated with nephrectomy. The study cohort was sub-divided into metastatic (n = 54, M1 at diagnosis or recurrence after surgery), high metastatic risk/HMR (n = 85, N1, T3-4, T2G3/4, T1G4) or low metastatic risk/LMR (n = 140, absence of these features) subsets. 3D primary tumor segmentation of arterial contrast CT scans was performed by trained investigators. Features were extracted using pyRadiomics 2.2.0 (n = 839) with gray value and voxel size normalization. For random forest (RF) model training, the cohort was randomly split into training (75%) and validation (25%) sets. Results: Multidimensional clustering of radiomic features by t-SNE analysis showed that metastatic and HMR tumors predominantly cluster together, while LMR tumors cluster separately. Consistent with this, there were no differentially regulated radiomic features (DR-features) between HMR and metastatic tumors. In contrast, we identified 26 DR-features (adjusted p-value &lt; 0.05) between presumed-metastatic (n = 139, HMR and metastatic tumors) and LMR tumors, which were then used as input to a RF binary classifier. In the training set, the trained classifier discriminated between presumed-metastatic and LMR tumors with bootstrapped AUC = 0.81. In the validation set, the classifier discriminated subsets with AUC = 0.80. Conclusions: High-risk and metastatic tumors have similar radiomic properties, suggesting common biology driving metastasis in RCC. We propose a novel radiomic classifier that accurately distinguishes between presumed-metastatic and low-risk tumors. Further work will assess whether this tool can identify patients with micrometastatic disease at diagnosis, who may benefit from adjuvant therapy or closer, long-term surveillance.

Abstract Duchenne muscular dystrophy (DMD) is a lethal X-linked disorder that results from mutations in the dystrophin gene causing necrosis, inflammation and ultimately fibrosis. In DMD, muscle macrophages are multifaceted effector immune cells that regulate multiple pathogenic processes, including myofiber injury, inflammation and fibrosis, but also promote regeneration. Previously, we demonstrated that regulatory T cells (Tregs) ameliorated the severity of muscular dystrophy by regulating M1-like and M2-like macrophage activation, and more recently, a subset of undefined macrophages. Herein, we used single-cell RNA sequencing (scRNAseq) approaches to define the phenotypic complexity and transcriptional profile of muscle macrophages in healthy and dystrophic muscle. A T-distributed Stochastic Neighbor Embedding (t-SNE) analysis of gene expression data revealed that muscle macrophages clustered into six novel populations defined by unique transcriptional programs. We observed that a population expressing high levels of galectin-3 was absent in wildtype and dystrophic muscle before the onset of disease, but was increased at the onset of acute pathology and further expanded when Tregs were depleted in the mdx mouse model of DMD. In light of emerging evidence in the literature suggesting a role for galectin-3 and Tregs in fibrosis, we performed in vivo functional assays to underscore the role of the Treg-Macrophages axis in the control of fibrosis in DMD. Our data support a new paradigm, in which unique transcriptional programs define novel macrophage populations likely adapted to the diseased muscle milieu, and suggest that therapies focused on augmenting Tregs function in dystrophic muscle may be beneficial to treat fibrosis.

Abstract Pregnancy is a significant immunological event, which has been shown to induce changes in maternal T cells. Studies examining maternal T cell responses in pregnancy have largely been conducted using mice that express transgenic “surrogate” fetal antigens, and have focused mainly on divided alloreactive T cells. Yet, maternal T cells may exhibit a broader range of responses during pregnancy some of which may not rely upon TCR signaling or cellular proliferation. To investigate the full spectrum of endogenous maternal T cell responses in natural pregnancy, we used the Nur77 mouse model, which allows for a graded assessment of TCR signaling using a GFP reporter. Splenocytes from female CD45.2 Nur77 mice were labeled with CellTrace Violet (CTV) to detect cellular division and transferred into female CD45.1 congenic mice. These host females were then mated with either syngeneic CD45.1 or allogeneic Balb/c males or left naive non-mated controls. Splenocytes were collected on either embryonic day 10 or 18 (E10 or E18) and analyzed using flow cytometry. t-SNE analysis revealed few phenotypic differences between the experimental groups at E10 with significantly increased separation by E18, which was most evident in the CD4 compartment. Interestingly, this separation was driven primarily by changes in the expression of selectins (i.e. downregulation of CD62L) and cytokine receptors (i.e. increased expression of CD25) as opposed to changes in TCR signaling (GFP) or cellular division (CTV). Accordingly, we conclude that maternal T cell responses in pregnancy are significantly shaped by non-antigen driven signals. The functional impact of these non-antigen driven changes in cellular phenotype are unknown and currently under investigation.

Abstract Background We studied mechanisms of therapeutic efficacy of the heterodimeric IL-15 (hetIL-15) of human, macaque or mouse in several mouse tumor models, including the murine EO771 orthotopic breast cancer model in syngeneic C57BL/6 mice. Methods The effects of hetIL-15 on immune cells were compared in tumors, lymph nodes and spleen by flow cytometry, immunohistochemistry (IHC), gene expression profiling (Nanostring) and metabolic profiling of tumor infiltrating T cells. Results hetIL-15 administration resulted in tumor regression in 40% of the treated animals and resistance to tumor rechallenge, inducing long term immunological memory. Activated NK and CD8+ T cells were increased in hetIL-15 treated tumors. hetIL-15 administration resulted in increased intratumoral infiltration of conventional Dendritic cells (cDC1) and also induced a new population of intratumoral DC, (F4/80high DCs), which were correlated negatively with the tumor size. Phenotypic profiling of F4/80high DCs identified expression of several cDC1 specific markers, including CD103, IRF8 and XCR1. tSNE analysis clustered F4/80high DCs close to cDC1s and macrophages. Conclusion Complete eradication of EO771 tumors by hetIL-15 treatment correlates with different tumor infiltrating immune cell types. We propose that the anti-cancer activity of hetIL-15 in primary tumors is orchestrated by the interplay of CD8+T cells, cDC1 and a novel subset of antigen presenting cells with a distinct phenotypic profile. Effective hetIL-15 treatment leads to development of long term anti-tumor memory. hetIL-15 supported a favorable metabolic profile of intratumoral effector lymphocytes, important for their function.

Lung regeneration occurs in a variety of adult mammals after surgical removal of one lung (pneumonectomy). Previous studies of murine post-pneumonectomy lung growth have identified regenerative “hotspots” in subpleural alveolar ducts; however, the cell-types participating in this process remain unclear. To identify the single cells participating in post-pneumonectomy lung growth, we used laser microdissection, enzymatic digestion and microfluidic isolation. Single-cell transcriptional analysis of the murine alveolar duct cells was performed using the C1 integrated fluidic circuit (Fluidigm) and a custom PCR panel designed for lung growth and repair genes. The multi-dimensional data set was analyzed using visualization software based on the tSNE algorithm. The analysis identified 6 cell clusters; 1 cell cluster was present only after pneumonectomy. This post-pneumonectomy cluster was significantly less transcriptionally active than 3 other clusters and may represent a transitional cell population. A provisional cluster identity for 4 of the 6 cell clusters was obtained by embedding bulk transcriptional data into the tSNE analysis. The transcriptional pattern of the 6 clusters was further analyzed for genes associated with lung repair, matrix production, and angiogenesis. The data demonstrated that multiple cell-types (clusters) transcribed genes linked to these basic functions. We conclude that the coordinated gene expression across multiple cell clusters is likely a response to a shared regenerative microenvironment within the subpleural alveolar ducts.

Exhausted phenotype of follicular CD8 T cells in CVID

In this paper, we propose Gaussian Process (GP) sound synthesis. A GP is used to sample random continuous functions, which are then used for wavetable or waveshaping synthesis. The shape of the sampled functions is controlled with the kernel function of the GP. Sampling multiple times from the same GP generates perceptually similar but non-identical sounds. Since there are many ways to choose the kernel function and its parameters, an interface aids the user in sound selection. The interface is based on a two-dimensional visualization of the sounds grouped by their similarity as judged by a t-SNE analysis of their Mel Frequency Cepstral Coefficient (MFCC) representations.

HIF-1 (hypoxia-inducible factor 1) signaling played a vital role in HCC (hepatocellular carcinoma) prognosis. We aimed to establish an accurate risk scoring system for HCC prognosis prediction and treatment guidance. 424 samples from TCGA (The Cancer Genome Atlas) and 445 samples from GSE14520 dataset were included as the derivation and validation cohort, respectively. In the derivation cohort, prognostic relevant signatures were selected from sixteen HIF-1 related genes and LASSO regression was adopted for model construction. Tumor-infiltrating immune cells were calculated using CIBERSORT algorithm. HIF-1 signaling significantly increased in HCC samples compared with normal tissues. Scoring system based on SLC2A1, ENO1, LDHA and GAPDH exhibited a continuous predictive ability for OS (overall survival) in HCC patients. PCA and t-SNE analysis confirmed a reliable clustering ability of risk score in both cohorts. Patients were classified into high-risk and low-risk groups and the survival outcomes between the two groups showed significant differences. In the derivation cohort, Cox regression indicated the scoring system was an independent predictor for OS, which was validated in the validation cohort. Different infiltrating immune cells fraction and immune scores were also observed in different groups. Herein, a novel integrated scoring system was developed based on HIF-1 related genes, which would be conducive to the precise treatment of patients.

Abstract Purpose: Inflammatory breast cancer (IBC) is an aggressive variant of breast cancer characterized by visible skin symptoms on the breast at the time of presentation. We demonstrated mesenchymal stem cells (MSC) promote skin symptoms in the SUM149 xenograft model. Here, we extend this finding to IBC patient-derived xenograft (PDX) lines and performed RNA-sequencing analysis of tumors from IBC patient-derived xenograft (PDX) lines by presence of skin symptoms. We hypothesize host MSC variation may explain sporadic skin symptoms in IBC PDX tumors across generations. Methods: MSC were co-injected at the time of tumor initiation in PDX tumors (IBC-3, 5, 6, and 7) and skin symptoms assessed visually and at the time of gross resection. A priori analysis plan was to group PDX by treatment group irrespective of PDX line. IBC PDX models spontaneously but inconsistently develop skin symptoms after multiple passages. RNA sequencing was performed on IBC-PDX tumors (n=33) and non-IBC PDX tumors (n=12) across generations from IBC PDX lines 5, 6, and 7 and non-IBC PDX lines 2, 3, and 4. Four samples were discarded due to poor quality. TSNE was applied using the top differential expressed genes. Xenome analysis was performed to examine host stromal expression by skin involvement. Samples from PDX lines 5 were compared with lines 6 and 7 and the top differentially expressed genes were compared with the Molecular Signatures Databases (MSigDB). CIBERSORT was performed to deconvolute cell type composition. Results: Co-injecting MSC with PDX transplant increased skin symptoms (57% of PDX animals (cohort from IBC5, IBC6, IBC7 and non-IBC4) with MSC vs. 0% without (cohort from IBC5, IBC6, IBC7, non-IBC2 and non-IBC4) (P = 0.0007). TSNE analysis of IBC PDX split samples into three subgroups distinguished by PDX lines and skin invasion. Xenome was used to split reads from human or mouse: samples show high percentage of reads coming from human (mean±SD, 97.44%± 8.98%). Seventy-four genes show FDR ≤ 0.2 between PDX line 5 and line 6 and comparing these top 74 genes with MSigDb for interaction revealed pathways such as GPCR, signal transduction, and sensory perception. CIBERSORT revealed no enrichment for MSC in IBC PDX with skin invasion, however significant differences in CD8 expression in these immunosuppressed PDX models casts doubt on the robustness of this finding. Conclusions: Exogenous MSC enhance skin symptoms in a cohort of IBC and non-IBC PDX. Skin symptoms segregate IBC PDX tumors by expression suggesting specific biology drives this presentation. However, we are unable to demonstrate evidence for spontaneous MSC signaling in these models. Future work is warranted to link top skin symptom related signaling to stromal signals and unravel the unexpected CD8 signaling identified in these models. Citation Format: Omar Rahal, Jie Yang, Li Li, Richard Larson, Steven Van Laere, Naoto Ueno, Erik Sulman, Wendy Woodward. Transcriptome analysis of patient derived xenograft mouse models of inflammatory breast cancer to gain insights into the contribution of tumor microenvironment [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P6-06-08.

MicroRNA miR-155 is required for expansion of regulatory T cells to mediate robust pregnancy tolerance in mice.

It is estimated that 50% of men and 25% of women worldwide suffer from hair loss, and therefore it is of great significance to investigate the molecular pathways driving hair follicle de novo morphogenesis. However, due to high cellular heterogeneity and the asynchronous development of hair follicles, our current understanding of the molecular mechanisms involved in follicle development remains limited.Methods: Single-cell suspensions from the dorsal skin of E13.5 (induction stage), E16.5 (organogenesis) fetal mice, and newborn mice (cytodifferentiation stage, postnatal day 0, P0) were prepared for unbiased single-cell RNA sequencing. To delineate the single-cell transcriptional landscape during hair follicle de novo morphogenesis, we performed t-distributed Stochastic Neighbor Embedding (tSNE), pseudotime cell trajectory inference, and regulon enrichment analysis to dissect cellular heterogeneity and reveal the molecular pathways underlying major cell type cell fate decisions. To validate our analysis, we further performed immunohistochemistry analysis of the key molecules involved during hair follicle morphogenesis. Meanwhile, intercellular communication between different cell populations was inferred based on a priori knowledge of ligand-receptor pairs.Results: Based on tSNE analysis, we identified 14 cell clusters from skin tissue and delineated their cellular identity from specific gene expression profiles. By using pseudotime ordering analysis, we successfully constructed the epithelium/dermal cell lineage differentiation trajectory. For dermal cell lineage, our analysis here recapitulated the dynamic gene expression profiles during dermal condensate (DC) cell fate commitment and delineated the heterogeneity of the different dermal papilla (DP) cell populations during in utero hair follicle development. For the epithelium cell lineage, our analysis revealed the dynamic gene expression profiles of the underappreciated matrix, interfollicular epidermis (IFE), hair shaft and inner root sheath (IRS) cell populations. Furthermore, single-cell regulatory network inference and clustering analysis revealed key regulons during cell fate decisions. Finally, intercellular communication analysis demonstrated that strong intercellular communication was involved during early hair follicle development.Conclusions: Our findings here provide a molecular landscape during hair follicle epithelium/dermal cell lineage fate decisions, and recapitulate the sequential activation of core regulatory transcriptional factors (TFs) in different cell populations during hair follicle morphogenesis. More importantly, our study here represents a valuable resource for understanding the molecular pathways involved during hair follicle de novo morphogenesis, which will have implications for future hair loss treatments.

Although the attention-based speech recognition has achieved promising performances, the specific explanation of the intermediate representations remains a black box theory. In this paper, we use the method to visually show and explain continuous encoder outputs. We propose a human-intervened force alignment method to obtain labels for t-distributed stochastic neighbor embedding (t-SNE), and use them to better understand the attention mechanism and the recurrent representations. In addition, we combine t-SNE and canonical correlation analysis (CCA) to analyze the training dynamics of phones in the attention-based model. Experiments are carried on TIMIT and WSJ respectively. The aligned embeddings of the encoder outputs could form sequence manifolds of the ground truth labels. Figures of t-SNE embeddings visually show what representations the encoder shaped into and how the attention mechanism works for the speech recognition. The comparisons between different models, different layers, and different lengths of the utterance show that manifolds are clearer in the shape when outputs are from the deeper layer of the encoder, the shorter utterance, and models with better performances. We also observe that the same symbols from different utterances tend to gather at similar positions, which proves the consistency of our method. Further comparisons are taken between different epochs of the model using t-SNE and CCA. The results show that both the plosive and the nasal/flap phones converge quickly, while the long vowel phone converge slowly.

ABSTRACT The induction of tumor-targeted, cytotoxic T lymphocytes has been recognized as a key component to successful immunotherapy. DPX-based treatment was previously shown to effectively recruit activated CD8+ T cells to the tumor. Herein, we analyze the unique phenotype of the CD8+ T cells recruited into the tumor in response to DPX-based therapy, and how combination with checkpoint inhibitors impacts T cell response. C3-tumor-bearing mice were treated with cyclophosphamide (CPA) for seven continuous days every other week, followed by DPX treatment along with anti-CTLA-4 and/or anti-PD-1. Efficacy, immunogenicity, and CD8+ T cells tumor infiltration were assessed. The expression of various markers, including checkpoint markers, peptide specificity, and proliferation and activation markers, was determined by flow cytometry. tSNE analysis of the flow data revealed a resident phenotype of CD8+ T cells (PD-1+TIM-3+CTLA-4+) within untreated tumors, whereas DPX/CPA treatment induced recruitment of a novel population of CD8+ T cells (PD-1+TIM-3+CTLA-4−) within tumors. Combination of anti-CTLA-4 (ipilimumab) with DPX/CPA versus DPX/CPA alone significantly increased survival and inhibition of tumor growth, without changing overall systemic immunogenicity. Addition of checkpoint inhibitors did not significantly change the phenotype of the newly recruited cells induced by DPX/CPA. Yet, anti-CTLA-4 treatment in combination with DPX/CPA enhanced a non-antigen specific response within the tumor. Finally, the tumor-recruited CD8+ T cells induced by DPX/CPA were highly activated, antigen-specific, and proliferative, while resident phenotype CD8+ T cells, seemingly initially exhausted, were reactivated with combination treatment. This study supports the potential of combining DPX/CPA with ipilimumab to further enhance survival clinically.

High-throughput RNA sequencing (RNA-Seq) has transformed the ecophysiological assessment of individual plankton species and communities. However, the technology generates complex data consisting of millions of short-read sequences that can be difficult to analyze and interpret. New bioinformatics workflows are needed to guide experimentation, environmental sampling, and to develop and test hypotheses. One complexity-reducing tool that has been used successfully in other fields is “t-distributed Stochastic Neighbor Embedding” (t-SNE). Its application to transcriptomic data from marine pelagic and benthic systems has yet to be explored. The present study demonstrates an application for evaluating RNA-Seq data using previously published, conventionally analyzed studies on the copepods Calanus finmarchicus and Neocalanus flemingeri. In one application, gene expression profiles were compared among different developmental stages. In another, they were compared among experimental conditions. In a third, they were compared among environmental samples from different locations. The profile categories identified by t-SNE were validated by reference to published results using differential gene expression and Gene Ontology (GO) analyses. The analyses demonstrate how individual samples can be evaluated for differences in global gene expression, as well as differences in expression related to specific biological processes, such as lipid metabolism and responses to stress. As RNA-Seq data from plankton species and communities become more common, t-SNE analysis should provide a powerful tool for determining trends and classifying samples into groups with similar transcriptional physiology, independent of collection site or time.

Uterine neoplasms comprise a broad spectrum of lesions, some of which may pose a diagnostic challenge even to experienced pathologists. Recently, genome-wide DNA methylation-based classification of central nervous system tumors has been shown to increase diagnostic precision in clinical practice when combined with standard histopathology. In this study, we describe DNA methylation patterns of a diverse set of uterine neoplasms and test the applicability of array-based DNA methylation profiling. A multicenter cohort including prototypical epithelial and mesenchymal uterine neoplasms was collected. Tumors were subject to pathology review and array-based DNA methylation profiling (Illumina Infinium HumanMethylation450 or EPIC [850k] BeadChip). Methylation data were analyzed by unsupervised hierarchical clustering and t-SNE analysis. After sample retrieval and pathology review the study cohort consisted of 49 endometrial carcinomas (EC), 5 carcinosarcomas (MMMT), 8 uterine leiomyomas (ULMO), 7 uterine leiomyosarcomas (ULMS), 15 uterine tumor resembling ovarian sex cord tumors (UTROSCT), 17 low-grade endometrial stromal sarcomas (LGESS) and 9 high-grade endometrial stromal sarcomas (HGESS). Analysis of methylation data identified distinct methylation clusters, which correlated with established diagnostic categories of uterine neoplasms. MMMT clustered together with EC, while ULMO, ULMS and UTROSCT each formed distinct clusters. The LGESS cluster differed from that of HGESS, and within the branch of HGESS, we observed a notable subgrouping of YWHAE- and BCOR-rearranged tumors. Herein, we describe distinct DNA methylation signatures in uterine neoplasms and show that array-based DNA methylation analysis holds promise as an ancillary tool to further characterize uterine neoplasms, especially in cases which are diagnostically challenging by conventional techniques.