Explore the vast range of titles available.

The specificity and sensitivity of a simple non-invasive multi-cancer screening method in detecting breast, lung, prostate, and colorectal cancer in breath samples were evaluated in a double-blind study. Breath samples of 1386 participants (59.7% males, median age 56.0 years) who underwent screening for cancer using gold-standard screening methods, or a biopsy for a suspected malignancy were collected. The samples were analyzed using a bio-hybrid platform comprising trained detection canines and artificial intelligence tools. According to cancer screening/biopsy results, 1048 (75.6%) were negative for cancer and 338 (24.4%) were positive. Among the 338 positive samples, 261 (77.2%) were positive for one of the four cancer types that the bio-hybrid platform was trained to detect, with an overall sensitivity and specificity of 93.9% (95% confidence interval [CI] 90.3-96.2%) and 94.3% (95% CI 92.7%-95.5%), respectively. The sensitivity of each cancer type was similar; breast: 95.0% (95% CI 87.8-98.0%), lung: 95.0% (95% CI 87.8-98.0%), colorectal: 90.0% (95% CI 74.4-96.5%), prostate: 93.0% (95% CI 84.6-97.0%). The sensitivity of 14 other malignant tumors that the bio-hybrid platform was not trained to detect, but identified, was 81.8% (95% CI 71.8%-88.8%). Early cancer (0–2) detection sensitivity was 94.8% (95% CI 91.0%-97.1%). This bio-hybrid multi-cancer screening platform demonstrated high sensitivity and specificity and enables early-stage cancer detection.

This paper presents a new approach for trees-based regression, such as simple regression tree, random forest and gradient boosting, in settings involving correlated data. We show the problems that arise when implementing standard trees-based regression models, which ignore the correlation structure. Our new approach explicitly takes the correlation structure into account in the splitting criterion, stopping rules and fitted values in the leaves, which induces some major modifications of standard methodology. The superiority of our new approach over trees-based models that do not account for the correlation is supported by simulation experiments and real data analyses.

This paper proposes parametric and non-parametric hypothesis testing algorithms for detecting anisotropy -- rotational variance of the covariance function in random fields. Both algorithms are based on resampling mechanisms, which enable avoiding relying on asymptotic assumptions, as is common in previous algorithms. The algorithms' performance is illustrated numerically in simulation experiments and on real datasets representing a variety of potential challenges.

K-fold cross-validation (CV) with squared error loss is widely used for evaluating predictive models, especially when strong distributional assumptions cannot be taken. However, CV with squared error loss is not free from distributional assumptions, in particular in cases involving non-i.i.d. data. This paper analyzes CV for correlated data. We present a criterion for suitability of standard CV in presence of correlations. When this criterion does not hold, we introduce a bias corrected cross-validation estimator which we term \\(CV_c,\\) that yields an unbiased estimate of prediction error in many settings where standard CV is invalid. We also demonstrate our results numerically, and find that introducing our correction substantially improves both, model evaluation and model selection in simulations and real data studies.

Common model selection criteria, such as \\(AIC\\) and its variants, are based on in-sample prediction error estimators. However, in many applications involving predicting at interpolation and extrapolation points, in-sample error cannot be used for estimating the prediction error. In this paper new prediction error estimators, \\(tAI\\) and \\(Loss(w_{t})\\) are introduced. These estimators generalize previous error estimators, however are also applicable for assessing prediction error in cases involving interpolation and extrapolation. Based on the prediction error estimators, two model selection criteria with the same spirit as \\(AIC\\) are suggested. The advantages of our suggested methods are demonstrated in simulation and real data analysis of studies involving interpolation and extrapolation in a Linear Mixed Model framework.

The STIL protein is essential for centriole replication and for the non-templated, de novo centriole biogenesis that is required for mammalian embryogenesis. Here we performed quantitative biophysical and structural analysis of the central short coiled coil domain (CCD) of STIL that is critical for its function. Using biophysical, biochemical and cell biology approaches, we identified the specific residues in the CCD that mediate the oligomerization, centrosomal localization and protein interactions of STIL. We characterized the structural properties of the coiled coil peptide using circular dichroism spectroscopy and size exclusion chromatography. We identified two regions in this domain, containing eight hydrophobic residues, which mediate the coiled coil oligomerization. Mutations in these residues destabilized the coiled coil thermodynamically but in most cases did not affect its secondary structure. Reconstituting mouse embryonic fibroblasts lacking endogenous Stil, we show that STIL oligomerization mediated by these residues is not only important for the centrosomal functions of STIL during the canonical duplication process but also for de-novo formation of centrosomes.