Search

Myotonic dystrophy type 1 is a complex disease caused by a genetically unstable CTG repeat expansion in the 3′-untranslated region of the DMPK gene. Age-dependent, tissue-specific somatic instability has confounded genotype–phenotype associations, but growing evidence suggests that it also contributes directly toward disease progression. Using a well-characterized clinical cohort of DM1 patients from Costa Rica, we quantified somatic instability in blood, buccal cells, skin and skeletal muscle. Whilst skeletal muscle showed the largest expansions, modal allele lengths in skin were also very large and frequently exceeded 2000 CTG repeats. Similarly, the degree of somatic expansion in blood, muscle and skin were associated with each other. Notably, we found that the degree of somatic expansion in skin was highly predictive of that in skeletal muscle. More importantly, we established that individuals whose repeat expanded more rapidly than expected in one tissue (after correction for progenitor allele length and age) also expanded more rapidly than expected in other tissues. We also provide evidence suggesting that individuals in whom the repeat expanded more rapidly than expected in skeletal muscle have an earlier age at onset than expected (after correction for the progenitor allele length). Pyrosequencing analyses of the genomic DNA flanking the CTG repeat revealed that the degree of methylation in muscle was well predicted by the muscle modal allele length and age, but that neither methylation of the flanking DNA nor levels of DMPK sense and anti-sense transcripts could obviously explain individual- or tissue-specific patterns of somatic instability.

Heatmap showing fRMA level gene data from Affymetrix subsets in training and validation datasets.

Supplementary methods and results

Validation of CMS assay at the research molecular diagnostic laboratory. Each sample classified in to CMS subtype 1-4 or mixed (M) subtype.

3A. Classification accuracy of CMS1 samples on training data set V1 using 4-fold cross validation as function of number of genes.3B.able-3B: Classification accuracy of CMS1 samples on validation data set (V2) as function of number of genes.3C. Classification accuracy CMS1 samples with out of sample (V2o) prediction set as function of number of genes.3D. Classification accuracy CMS1 samples with TCGA dataset (RNA-seq subset) as function of number of genes. 3E. Classification accuracy CMS1 samples with Affymetrix set (V2a) as function of number of genes.

Classification accuracy on validation data set, the entire set (V2), out-of-sample prediction (V2o), Affymetrix subset (V2a), and TCGA RNAseq subset (TCGA) for various methods as function of number of genes.

Purpose:Consensus molecular subtyping (CMS) of colorectal cancer has potential to reshape the colorectal cancer landscape. We developed and validated an assay that is applicable on formalin-fixed, paraffin-embedded (FFPE) samples of colorectal cancer and implemented the assay in a Clinical Laboratory Improvement Amendments (CLIA)-certified laboratory.Experimental Design:We performed an in silico experiment to build an optimal CMS classifier using a training set of 1,329 samples from 12 studies and validation set of 1,329 samples from 14 studies. We constructed an assay on the basis of NanoString CodeSets for the top 472 genes, and performed analyses on paired flash-frozen (FF)/FFPE samples from 175 colorectal cancers to adapt the classifier to FFPE samples using a subset of genes found to be concordant between FF and FFPE, tested the classifier's reproducibility and repeatability, and validated in a CLIA-certified laboratory. We assessed prognostic significance of CMS in 345 patients pooled across three clinical trials.Results:The best classifier was weighted support vector machine with high accuracy across platforms and gene lists (>0.95), and the 472-gene model outperforming existing classifiers. We constructed subsets of 99 and 200 genes with high FF/FFPE concordance, and adapted FFPE-based classifier that had strong classification accuracy (>80%) relative to “gold standard” CMS. The classifier was reproducible to sample type and RNA quality, and demonstrated poor prognosis for CMS1–3 and good prognosis for CMS2 in metastatic colorectal cancer (P < 0.001).Conclusions:We developed and validated a colorectal cancer CMS assay that is ready for use in clinical trials, to assess prognosis in standard-of-care settings and explore as predictor of therapy response.

Nanostring CMS Classifier Performance: Performance of Nanostring 100 gene CMS classifier applied to FFPE samples (6A and 6C) and FF samples (6B and 63D). The top two figures plot 4-class accuracy vs. classification confidence, with the dots marking individual samples either correctly (1.0) or incorrectly (0.0), with color indicat-ing correct CMS. The line contains a generalized additive model (GAM) fit to these data with 95% pointwise confidence bands, and demonstrates that samples classified with greater confidence were more likely to be cor-rectly classified. Figure 6C and 6D plot 4-class accuracy vs. RNA quality, defined as 0nt (FFPE) or RIN (FF). Note that there is little if any association of CMS accuracy with RNA quality, suggesting that the perfor-mance of classifier is robust to RNA quality in this study.

Classification accuracy of different modelling strategies on training data set (V1) using 4-fold cross validation.

Comparison of Classifiers. Comparison of random forest (RF) and single sample predictors (SSP) in Guinney et al. (2015) paper with our wSVM classifier with 472 genes, either forcing a single CMS call for each sample (all calls) or only when classified with high confidence

5A. Classification accuracy of CMS 3 samples on training data set V1 using 4-fold cross validation as function of number of genes. 5B.Classification accuracy of CMS 3 samples on validation data set (V2) as function of number of genes. 5C. Classification accuracy CMS3 samples with out of sample prediction set (V2o) as function of number of genes. 5D. Classification accuracy CMS 3 samples with TCGA dataset (RAN-seq subset) as function of number of genes. 5E. Classification accuracy CMS3 samples with Affymetrix dataset (V2a) as function of number of genes.

6A. Classification accuracy of CMS 4 samples on training data set V1 using 4-fold cross validation as function of number of genes. 6B. Classification accuracy of CMS 4 samples on validation data set (V2) as function of number of genes. 6C. Classification accuracy CMS 4 samples with out of sample prediction set (V2o) as function of number of genes. 6D. Classification accuracy CMS 3 samples with TCGA dataset (RAN-seq subset) as function of number of genes. 6E. Classification accuracy of CMS 3 samples with Affymetrix data set (V2a) as function of number of genes.

Sample-wise and Gene-wise correlation of FF/FFPE paired Samples: Scatterplot of gene-specific Spearman correlation of FF/FFPE computed separately for CRCSC samples and non-CRCSC samples , with red and blue marking the overall top 100 and top 200 genes, respectively, in terms of FF/FFPE correlation computed across all samples. Note the high level of agreement across batches that genes with high FF/FFPE correlations for one batch tended to also have high FF/FFPE correlation in the other batch, suggesting high FF/FFPE correla-tion is a characteristic of the gene/probe set.

4A. Classification accuracy of CMS2 samples on training data set V1 using 4-fold cross validation as function of number of genes.4B. Classification accuracy of CMS2 samples on validation data set (V2) as function of number of genes.4C. Classification accuracy CMS2 samples with out of sample(V2o) prediction set as function of number of genes. 4D. Classification accuracy CMS2 samples with TCGA dataset (RNA-seq subset) as function of number of genes.4E. Classification accuracy CMS2 samples with Affymentrix dataset (V2a) as function of number of genes.

Algorithm showing strategy to de-velop the Nanostring based FFPE CMS classifier using Affymetrix gene expression data. FF: flash frozen, FFPE: Formalin fixed par-affin embedded, NFFPE: Nanos-tring FFPE, AFF: Flash frozen Affymetrix, GES: Gene expres-sion score, CRCSC: Colorectal cancer subtyping consortium

Classification Confidence of Affymetrix Validation Samples vs. Classification Accuracy of 472-gene wSVM clas-sifier: Distribution of classification confidence (�i) across 713 validation samples (top), and classification accura-cy (1=correct, 0=incorrect) as a function of classification confidence (�i), with fitted loess curve and 95% confidence interval.

Classification accuracy for training data V1 (based on 4-fold cross val-idation, 2A) and the full validation data V2 (2B) for various methods as function of number

In this study, wind load on a typical jack-up is calculated through CFD simulation and measured through wind tunnel tests for operational condition (OC) and extreme condition (EC). The Reduced Order Method (ROM) model is built up based on high fidelity CFD simulation results. Two data assimilation models, ROM-based Gappy POD and Extended Kalman Filter (EKF) model, are developed to fuse the measured data from wind tunnel tests with the numerical results for jack-up wind load predictions. The outcomes show that both the EKF and gappy POD are able to predict the wind loads on the jack-up platform under OC and EC with satisfying accuracy, even with only 21 measurement data. The findings of this study provide an option for fusing the numerical simulation results with wind-tunnel measurements or operational data for accurately predicting the wind load on marine vessels and offshore platforms.

With the development of internet of things (IoT) technology, servitization of IoT device functions has become a trend. The cooperation between IoT devices can be equivalent to web service composition. However, current service composition approaches applied in the internet cannot work well in IoT environments due to weak adaptability, low accuracy, and poor time performance. This paper, based on service dependency graph, proposes a top-k QoS-optimal service composition approach suitable for IoT. It aims to construct the relationship between services by applying the service dependency model and to reduce the traversal space through effective filtering strategies. On the basis of a composition path traversal sequence, the generated service composition can be represented directly to avoid backtracking search. Meanwhile, the redundant services can be removed from the service composition with the help of dynamic programming. Experiments show that the approach can obtain the top-k QoS-optimal service composition and better time performance.

The mechanism by which anti-cancer immunity shapes early carcinogenesis of lung adenocarcinoma (ADC) is unknown. In this study, we characterize the immune contexture of invasive lung ADC and its precursors by transcriptomic immune profiling, T cell receptor (TCR) sequencing and multiplex immunofluorescence (mIF). Our results demonstrate that anti-tumor immunity evolved as a continuum from lung preneoplasia, to preinvasive ADC, minimally-invasive ADC and frankly invasive lung ADC with a gradually less effective and more intensively regulated immune response including down-regulation of immune-activation pathways, up-regulation of immunosuppressive pathways, lower infiltration of cytotoxic T cells (CTLs) and anti-tumor helper T cells (Th), higher infiltration of regulatory T cells (Tregs), decreased T cell clonality, and lower frequencies of top T cell clones in later-stages. Driver mutations, chromosomal copy number aberrations (CNAs) and aberrant DNA methylation may collectively impinge host immune responses and facilitate immune evasion, promoting the outgrowth of fit subclones in preneoplasia into dominant clones in invasive ADC. The evolution of immune landscape in lung adenocarcinoma (ADC) is largely unknown. Here the authors use a cohort of resected invasive lung ADC and its precursors and show a gradual increase of immunosuppression and decrease of anti-tumor response associated with specific genomic and epigenetic features.

Myotonic dystrophy type 1 (DM1) is a complex disease with a wide spectrum of symptoms. The exact relationship between mutant CTG repeat expansion size and clinical outcome remains unclear. DM1 congenital patients (CDM) inherit the largest expanded alleles, which are associated with abnormal and increased DNA methylation f lanking the CTG repeat. However, DNA methylation at the DMPK locus remains understudied. Its relationship to DM1 clinical subtypes, expansion size and age-at-onset is not yet completely understood. Using pyrosequencing-based methylation analysis on 225 blood DNA samples from Costa Rican DM1 patients, we determined that the size of the estimated progenitor allele length (ePAL) is not only a good discriminator between CDM and non-CDM cases (with an estimated threshold at 653 CTG repeats), but also for all DM1 clinical subtypes. Secondly, increased methylation at both CTCF sites upstream and downstream of the expansion was almost exclusively present in CDM cases. Thirdly, levels of abnormal methylation were associated with clinical subtype, age and ePAL, with strong correlations between these variables. Fourthly, both ePAL and the intergenerational expansion size were significantly associated with methylation status. Finally, methylation status was associated with ePAL and maternal inheritance, with almost exclusively maternal transmission of CDM. In conclusion, increased DNA methylation at the CTCF sites f lanking the DM1 expansion could be linked to ePAL, and both increased methylation and the ePAL could be considered biomarkers for the CDM phenotype. Muscular Dystrophy Association/[MDA200568]/MDA/Estados Unidos Universidad de Costa Rica/[742-A8-306]/UCR/Costa Rica National Cancer Institute/[P30 CA016672]/NCI/Estados Unidos UCR::Vicerrectoría de Investigación::Unidades de Investigación::Ciencias de la Salud::Instituto de Investigaciones en Salud (INISA)

For wireless sensor networks (WSN) with connection failure uncertainties, traditional minimum spanning trees are no longer a feasible option for selecting routes. Reliability should come first before cost since no one wants a network that cannot work most of the time. First, reliable route selection for WSNs with connection failure uncertainties is formulated by considering the top-k most reliable spanning trees (RST) from graphs with structural uncertainties. The reliable spanning trees are defined as a set of spanning trees with top reliabilities and limited tree weights based on the possible world model. Second, two tree-filtering algorithms are proposed: the k minimum spanning tree (KMST) based tree-filtering algorithm and the depth-first search (DFS) based tree-filtering algorithm. Tree-filtering strategy filters the candidate RSTs generated by tree enumeration with explicit weight thresholds and implicit reliability thresholds. Third, an innovative edge-filtering method is presented in which edge combinations that act as upper bounds for RST reliabilities are utilized to filter the RST candidates and to prune search spaces. Optimization strategies are also proposed for improving pruning capabilities further and for enhancing computations. Extensive experiments are conducted to show the effectiveness and efficiency of the proposed algorithms.

Wind tunnel test, Computational Fluid Dynamics (CFD) simulation, and Proper Orthogonal Decomposition (POD) based Reduced Order Model (ROM) are used for wind load prediction on a LNG carrier in this project. In order to train the model for high accurate predictions and provide wind load prediction for vessel operations with high confidence level, extended Kalman Filter (EKF) based data assimilation model is developed to fuse the experimental data and the ROM predicted data. Effects of sensor locations and sensor numbers on the data fusing are studied. Study results indicate that the single point data assimilation can help with sensor positioning and outlier data detection. Based on the standard errors at the sensor points, prediction accuracy with sorted multipoint data assimilation is higher than that with single point data assimilation.

Background: In the Phase III randomized trial (SWOG Lung-MAP S1400I non match trial), adding ipilimumab (I) to nivolumab (N) did not improve survival of patients with advanced, pretreated, immune checkpoint inhibitor naive lung squamous cell carcinoma (SCC). The I+N combination demonstrated superiority in patients with a tumor mutation burden (TMB) ≥10 mt/MB and PD-L1 expression Methods: RNA extracted from baseline FFPE samples from 66/252 patients enrolled in the S1400I trial was received from the SWOG bank with 28 samples considered inadequate due to low input RNA. 38 samples were run on the ncounter platform using the PanCancer Immune Profiling panel using the manufacturer’s instructions with the addition of Human Reference RNA control. Data were processed and normalized using nSolver. All samples passed the post run QC with no batch effect. Using a log-rank test on dichotomized normalized data, we performed time to event analysis to identify genes that correlated with overall survival (OS) and PFS (Progression Free Survival) for all 38 samples and for each arm (23 N and 15 I+N). We ran a Cox model using significant genes identified independently and in association with TMB≥10 and PD-L1 ≥5 thresholds identified in the clinical studies. TIMER and nSolver advanced analysis were used to infer immune cells that correlated to clinical outcomes. Results: We observed that BLNK, CD163, FCGR2A associated with increased OS (p Conclusion: Despite a small dataset, this analysis shows a potential advantage in PFS and OS with increased presence of immune cells including exhausted CD8+ T cells and genes associated with myeloid cells. Validation of these findings is warranted and may refine patient populations that would benefit from this combination strategy. Acknowledgments: Scientific and financial support: U10CA180888, U10CA180819, U24CA224285, U24CA224316, PACT, PPP and FNIH Citation Format: Dzifa Y. Duose, Jiexin Zhang, Mary Redman, Baili Zhang, Ethan Cerami, James Lindsay, Joyce Yu, Roshni Biswas, Stephen Van Nostrand, Radim Moravec, Rajyalakshmi Luthra, Gheath Al-Atrash, Karen Kelly, Roy Herbst, Ignacio Wistuba, Scott Gettinger, Lyudmila Bazhenova, J. Jack Lee, Jianjun Zhang, Cara Haymaker. Immune gene expression signatures associated clinical benefit from nivolumab and ipilimumab for previously treated patients with stage IV squamous cell lung cancer: An immune biomarker analysis of phase III SWOG LungMAP S1400I trial [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1974.

9046 Background: Immune checkpoint blockade (ICB) has become a standard pillar of treatment for lung cancer. However, only ̃20% of unselected patients can achieve durable clinical benefits. We performed immunogenomic profiling of tissue specimens from a randomized Phase III trial S1400I on metastatic lung squamous cell carcinoma (SCC) to evaluate if there were factors associated with better prognoses with ICB from single-agent versus combined targeting PD-1/CTLA-4 and evaluate if any differentiated between the treatments. Methods: We utilized FFPE tumor tissue submitted for Lung-MAP screening provided by the SWOG bank. SCC samples from 82 eligible patients treated with combined nivolumab+ipilimumab (N+I) or single agent nivolumab (N) were subjected to multiplex immunofluorescence (mIF, n = 82) and NanoString (ncounter PanCancer Immune Profiling Panel, n = 32). Cell density phenotypes (cells/mm2) were defined using image analysis of staining for cytokeratin, CD3, CD8, granzyme B, CD45RO, FOXP3, PD1, PD-L1, and CD68. Immunogenomic features were associated with response, PFS, and OS derived from data provided by the LungMap team to the CIDC portal. For statistical analyses, non-parametric tests were utilized to assess associations of cell phenotypes versus continuous or categorical variables, and log-rank test analysis was performed to identify cell phenotypes or genes correlated with survival. Results: In both arms higher densities of total CD3+CD45RO+ T cells ( P= 0.041), CD3+PD-1+ T cells ( P= 0.024) and CD3+CD8+PD-1 T cells in stroma ( P= 0.042) and CD3+CD8+GZMB+ T cells in the tumor compartment ( P= 0.011) were positively associated with PFS. In the N+I arm but not in the N arm, higher densities of CD3+CD8+GZMB+ T cells in the tumor compartment were associated with better PFS ( P= 0.015) and higher densities of stroma CD3+CD8-FOXP3+ T cells with worse OS. Spatial analysis showed that the presence of CD8+GZMB+ T cells close to malignant cells (median, ≤19.27 µm) was associated with better PFS ( P= 0.037) in N+I arm and cluster analysis showed low clustering of cells in TMB-high vs. TMB-low tumors (P < 0.01). Gene expression profiling demonstrated that myeloid infiltration, immune recruitment, and inflammation genes were associated with a positive clinical outcome ( P< 0.05). In both arms, BLNK, CD163, FCGR2A were associated with better OS ( P< 0.01), IRF1 and BLNK were associated with increased PFS ( P< 0.01). In the N+I arm but not in the N arm, we observed significantly higher CD45 immune cell scores, including CD8 T cells and neutrophils, in responders versus non-responders. Conclusions: Our findings suggest a potential advantage in PFS and OS with an increased presence of cytotoxic immune cells and genes associated with the recruitment and proliferation of these cell types before therapy.

Transformers are prone to metal corrosion due to long-term exposure in the external environment. It is necessary to detect corrosion in time to prolong their service life-time. Detecting metal corrosion based on deep learning is becoming a feasible method at present. However, because of the irregularity and detachability of metal corrosion, traditional annotation approach results in such problems as ambiguity and uncertainty in the annotating process. Thus, a novel hierarchical annotation approach is proposed in this paper. The rationale for this initiative is as follows: traditional annotation approach has been used to annotate a large area covering the range of corrosion, as long as the area is visually continuous and adjacent to the corrosion that cannot be clearly divided. The annotation result is recorded as training set A; secondly, in the annotating boxes from the first step, the areas with obvious and relatively independent features are re-annotated to form the second level of nested annotation. The annotation result is recorded as training set B. Finally, Faster R-CNN and YOLOv5 models were trained by A, B training sets respectively in the experiment. It can be concluded that the detection performance with hierarchical annotation approach proved better than traditional annotation approach for Faster R-CNN and YOLOv5 models.

To deal with low efficiency of existing Web service composition algorithms under dynamic environment, this paper proposes a Top-k QoS-optimal service composition algorithm based on QWSC-K static algorithm. The algorithm first considers the influence of dynamic environment on the existing service composition and service dependency graph, and classifies the changed services in dynamic environment. Moreover, different strategies are adopted to obtain the intermediate results of service composition for different categories of services. The influence of different service categories on the service dependency graph is analyzed to realize the updating of the service dependency graph. Finally, based on the updated service dependency graph and the intermediate results of the service composition, the adaptive incremental update can be applied to change the original service composition results, avoiding the time-consuming re-query process. Experiments show that the algorithm is efficient and accurate in dynamic environment.