Search

Medulloblastoma is the most common embryonic brain tumor in children. We investigated a cohort of 52 Asian medulloblastoma patients aged between 0 and 19 years old, who received surgical resections and post-resection treatments in the Taipei Medical University Hospital and the Taipei Veterans General Hospital. Genome-wide RNA sequencing was performed on fresh-frozen surgical tissues. These data were analyzed using the CIBERSORTx immune deconvolution software. Two external clinical and molecular datasets from United States (n = 62) and Canada (n = 763) were used to evaluate the transferability of the gene-signature scores across ethnic populations. The abundance of 13 genes, including DLL1, are significantly associated with overall survival (All Cox regression P < 0.001). A gene-signature score was derived from the deep transcriptome, capable of indicating patients' subsequent tumor recurrence (Hazard Ratio [HR] 1.645, confidence interval [CI] 1.337–2.025, P < 0.001) and mortality (HR 2.720, CI 1.798–4.112, P < 0.001). After the adjustment of baseline clinical factors, the score remains indicative of recurrence-free survival (HR 1.604, CI 1.292–1.992, P < 0.001) and overall survival (HR 2.781, CI 1.762–4.390, P < 0.001). Patients stratified by this score manifest not only distinct prognosis but also different molecular characteristics: Notch signaling ligands and receptors are comparatively overexpressed in patients with poorer prognosis, while tumor infiltrating natural killer cells are more abundant in patients with better prognosis. Additionally, immunohistochemical staining showed the DLL1 protein, a major ligand in the Notch signaling pathway, and the NCAM1 protein, a representative biomarker of natural killer cells, are present in the surgical tissues of patients of four molecular subgroups, WNT, SHH, Group 3 and Group 4. NCAM1 RNA level is also positively associated with the mutation burden in tumor (P = 0.023). The gene-signature score is validated successfully in the Canadian cohort (P = 0.009) as well as its three molecular subgroups (SHH, Group 3 and Group 4; P = 0.047, 0.018 and 0.040 respectively). In conclusion, pediatric medullablastoma patients can be stratified by gene-signature scores with distinct prognosis and molecular characteristics. Ligands and receptors of the Notch signaling pathway are overexpressed in the patient stratum with poorer prognosis. Tumor infiltrating natural killer cells are more abundant in the patient stratum with better prognosis. [ABSTRACT FROM AUTHOR]

In this article, we propose a class of partial deconvolution kernel estimators for the nonparametric regression function when some covariates are measured with error and some are not. The estimation procedure combines the classical kernel methodology and the deconvolution kernel technique. According to whether the measurement error is ordinarily smooth or supersmooth, we establish the optimal local and global convergence rates for these proposed estimators, and the optimal bandwidths are also identified. Furthermore, lower bounds for the convergence rates of all possible estimators for the nonparametric regression functions are developed. It is shown that, in both the super and ordinarily smooth cases, the convergence rates of the proposed partial deconvolution kernel estimators attain the lower bound. The Canadian Journal of Statistics 48: 535–560; 2020 © 2020 Statistical Society of Canada [ABSTRACT FROM AUTHOR]

Case studies are presented in this paper to demonstrate the use of the pressure/rate deconvolution-approach in estimating drainage areas for wells completed in some of the naturally fractured tight gas reservoirs of the Canadian Rockies foothills. These case studies demonstrate the application of deconvolution to two key carbonate-stratigraphical horizons in the area: the Triassic Baldonnel and the Permo-Carboniferous Taylor Flat formations. In these structural plays with significant areal formation-rock heterogeneity, the matrix-rock properties controlling the gas storativity are low, with porosity between 3 and 6%, causing low matrix-rock permeability (from 0.01 and 0.1 md). However, all of these formations have been thrusted, overturned, and subjected to reverse faulting. These diagenetic factors have created swarms of natural fractures that control flow rates and may define rock volumes connected to individual wells. In each well, a preproduction flow test was performed with the intent of ensuring acceptable flow rates and scoping facility design. At this stage of early development, initial-gas-in-place (IGIP) estimates were derived mainly from geophysical mapping, with plans to calibrate the IGIP number through the application of gas material balance, rate-transient analysis, and/or simple latetime rate decline. The rate-history data available in the early stage of production were integrated with pressure-buildup (PBU) data collected later in the production life of the well during annual or routine shut-in periods that were relatively short. Application of deconvolution in this paper is aimed at detecting early signs of pseudosteady-state pool depletion and estimating connected drainage volume. The deconvolution procedures help calibrate and/or reconcile geosciences-defined volumetric resource sizes, map remaining reserves, and help identify possible infill-drilling opportunities. [ABSTRACT FROM AUTHOR]