Cell survival following direct executioner-caspase activation.

Executioner-caspase activation has been considered a point-of-no-return in apoptosis. However, numerous studies report survival from caspase activation after treatment with drugs or radiation. An open question is whether cells can recover from direct caspase activation without pro-survival stress responses induced by drugs. To address this question, we engineered a HeLa cell line to express caspase-3 inducibly and combined it with a quantitative caspase activity reporter. While high caspase activity levels killed all cells and very low levels allowed all cells to live, doses of caspase activity sufficient to kill 15 to 30% of cells nevertheless allowed 70 to 85% to survive. At these doses, neither the rate, nor the peak level, nor the total amount of caspase activity could accurately predict cell death versus survival. Thus, cells can survive direct executioner-caspase activation, and variations in cellular state modify the outcome of potentially lethal caspase activity. Such heterogeneities may underlie incomplete tumor cell killing in response to apoptosis-inducing cancer treatments.

Comparing the reliability and predictive power of child, teacher, and guardian reports of noncognitive skills.

Children's noncognitive or socioemotional skills (e.g., persistence and self-control) are typically measured using surveys in which either children rate their own skills or adults rate the skills of children. For many purposes-including program evaluation and monitoring school systems-ratings are often collected from multiple perspectives about a single child (e.g., from both the child and an adult). Collecting data from multiple perspectives is costly, and there is limited evidence on the benefits of this approach. Using a longitudinal survey, this study compares children's noncognitive skills as reported by themselves, their guardians, and their teachers. Although reports from all three types of respondents are correlated with each other, teacher reports have the highest internal consistency and are the most predictive of children's later cognitive outcomes and behavior in school. The teacher reports add predictive power beyond baseline measures of Intelligence Quotient (IQ) for most outcomes in schools. Measures collected from children and guardians add minimal predictive power beyond the teacher reports.

Predictability of ecological and evolutionary dynamics in a changing world.

Ecological and evolutionary predictions are being increasingly employed to inform decision-makers confronted with intensifying pressures on biodiversity. For these efforts to effectively guide conservation actions, knowing the limit of predictability is pivotal. In this study, we provide realistic expectations for the enterprise of predicting changes in ecological and evolutionary observations through time. We begin with an intuitive explanation of predictability (the extent to which predictions are possible) employing an easy-to-use metric, predictive power PP(t). To illustrate the challenge of forecasting, we then show that among insects, birds, fishes and mammals, (i) 50% of the populations are predictable at most 1 year in advance and (ii) the median 1-year-ahead predictive power corresponds to a prediction R 2 of only 20%. Predictability is not an immutable property of ecological systems. For example, different harvesting strategies can impact the predictability of exploited populations to varying degrees. Moreover, incorporating explanatory variables, accounting for time trends and considering multivariate time series can enhance predictability. To effectively address the challenge of biodiversity loss, researchers and practitioners must be aware of the information within the available data that can be used for prediction and explore efficient ways to leverage this knowledge for environmental stewardship.

Bayesian hierarchical modeling in interim futility analysis for two parallel clinical trials.

Incorporating interim analysis into a trial design is gaining popularity in the field of confirmatory clinical trials, where two studies may be conducted in parallel (ie, twin studies) in order to provide substantial evidence per the requirement of FDA guidance. Interim futility analysis provides a chance to check for the "disaster" scenario when the treatment has a high probability to be not more efficacious than the control. Therefore, it is an efficient tool to mitigate risk of running a complete and expansive trial under such scenario. There is no agreement among trial designers that interim analysis should be based on individual study data or pooled data under the twin study scenario. In fact, it is a dilemma for most scientists when specifying the interim analysis strategy at the design stage as the true treatment effects of the twin studies are unknown no matter how similar they are intended to be. To address the issue, we developed a Bayesian hierarchical modeling method to allow dynamic data borrowing between twin studies and demonstrated a favorable characteristic of the new method over the separate and pooled analyses. We evaluated a wide spectrum of the heterogeneity hyperparameters and visualized its critical impact on the Bayesian model's characteristic. Based on the evaluation, we made a suggestion on the heterogeneity hyperparameter selection independent of any a priori knowledge. We also applied our method to a case study where predictive powers of different methods are compared.

Seamless phase 2/3 design for trials with multiple co-primary endpoints using Bayesian predictive power.

Seamless phase 2/3 design has become increasingly popular in clinical trials with a single endpoint. Trials that define success based on the achievement of all co-primary endpoints (CPEs) encounter the challenge of inflated type 2 error rates, often leading to an overly large sample size. To tackle this challenge, we introduced a seamless phase 2/3 design strategy that employs Bayesian predictive power (BPP) for futility monitoring and sample size re-estimation at interim analysis. The correlations among multiple CPEs are incorporated using a Dirichlet-multinomial distribution. An alternative approach based on conditional power (CP) was also discussed for comparison. A seamless phase 2/3 vaccine trial employing four binary endpoints under the non-inferior hypothesis serves as an example. Our results spotlight that, in scenarios with relatively small phase 2 sample sizes (e.g., 50 or 100 subjects), the BPP approach either outperforms or matches the CP approach in terms of overall power. Particularly, with n1 = 50 and ρ = 0, BPP showcases an overall power advantage over CP by as much as 8.54%. Furthermore, when the phase 2 stage enrolled more subjects (e.g., 150 or 200), especially with a phase 2 sample size of 200 and ρ = 0, the BPP approach evidences a peak difference of 5.76% in early stop probability over the CP approach, emphasizing its better efficiency in terminating futile trials. It's noteworthy that both BPP and CP methodologies maintained type 1 error rates under 2.5%. In conclusion, the integration of the Dirichlet-Multinominal model with the BPP approach offers improvement in certain scenarios over the CP approach for seamless phase 2/3 trials with multiple CPEs.

Associations and predictive power of dietary patterns on metabolic syndrome and its components.

BACKGROUND AND AIMS: Metabolic syndrome (MetS) defines important risk factors in the development of cardiovascular diseases and other serious health conditions. This study aims to investigate the influence of different dietary patterns on MetS and its components, examining both associations and predictive performance. METHODS AND RESULTS: The study sample included 10,750 participants from the seventh survey of the cross-sectional, population-based Tromsø Study in Norway. Diet intake scores were used as covariates in logistic regression models, controlling for age, educational level and other lifestyle variables, with MetS and its components as response variables. A diet high in meat and sweets was positively associated with increased odds of MetS and elevated waist circumference, while a plant-based diet was associated with decreased odds of hypertension in women and elevated levels of triglycerides in men. The predictive power of dietary patterns derived by different dimensionality reduction techniques was investigated by randomly partitioning the study sample into training and test sets. On average, the diet score variables demonstrated the highest predictive power in predicting MetS and elevated waist circumference. The predictive power was robust to the dimensionality reduction technique used and comparable to using a data-driven prediction method on individual food variables. CONCLUSIONS: The strongest associations and highest predictive power of dietary patterns were observed for MetS and its single component, elevated waist circumference.

Epistemic uncertainty in Bayesian predictive probabilities.

Bayesian predictive probabilities have become a ubiquitous tool for design and monitoring of clinical trials. The typical procedure is to average predictive probabilities over the prior or posterior distributions. In this paper, we highlight the limitations of relying solely on averaging, and propose the reporting of intervals or quantiles for the predictive probabilities. These intervals formalize the intuition that uncertainty decreases with more information. We present four different applications (Phase 1 dose escalation, early stopping for futility, sample size re-estimation, and assurance/probability of success) to demonstrate the practicality and generality of the proposed approach.

Reviewing and assessing existing meta-analysis models and tools.

Over the past few years, meta-analysis has become popular among biomedical researchers for detecting biomarkers across multiple cohort studies with increased predictive power. Combining datasets from different sources increases sample size, thus overcoming the issue related to limited sample size from each individual study and boosting the predictive power. This leads to an increased likelihood of more accurately predicting differentially expressed genes/proteins or significant biomarkers underlying the biological condition of interest. Currently, several meta-analysis methods and tools exist, each having its own strengths and limitations. In this paper, we survey existing meta-analysis methods, and assess the performance of different methods based on results from different datasets as well as assessment from prior knowledge of each method. This provides a reference summary of meta-analysis models and tools, which helps to guide end-users on the choice of appropriate models or tools for given types of datasets and enables developers to consider current advances when planning the development of new meta-analysis models and more practical integrative tools.

Comparison of VTE risk scores in guidelines for VTE diagnosis in nonsurgical hospitalized patients with suspected VTE.

BACKGROUND: The assessment of VTE likelihood with VTE risk scores is essential prior to imaging examinations during VTE diagnostic procedure. Little is known with respect to the disparity of predictive power for VTE diagnosis among VTE risk scores in guidelines for nonsurgical hospitalized patients with clinically suspected VTE. METHODS: A retrospective study was performed to compare the predictive power for VTE diagnosis among the Wells, Geneva, YEARS, PERC, Padua, and IMPROVE scores in the leading authoritative guidelines in nonsurgical hospitalized patients with suspected VTE. RESULTS: Among 3168 nonsurgical hospitalized patients with suspected VTE, VTE was finally excluded in 2733(86.3%) ones, whereas confirmed in 435(13.7%) ones. The sensitivity and specificity resulted from the Wells, Geneva, YEARS, PERC, Padua, and IMPROVE scores were (90.3%, 49.8%), (88.7%, 53.6%), (73.8%, 50.2%), (97.7%,16.9%), (80.9%, 44.0%), and (78.2%, 47.0%), respectively. The YI were 0.401, 0.423, 0.240, 0.146, 0.249, and 0.252 for the Wells, Geneva, YEARS, PERC, Padua, and IMPROVE scores, respectively. The C-index were 0.694(0.626-0.762), 0.697(0.623-0.772), 0.602(0.535-0.669), 0.569(0.486-0.652), 0.607(0.533-0.681), and 0.609(0.538-0.680) for the Wells, Geneva, YEARS, PERC, Padua, and IMPROVE scores, respectively. Consistency was significant in the pairwise comparison of Wells vs Geneva(Kappa 0.753, P = 0.565), YEARS vs Padua(Kappa 0.816, P = 0.565), YEARS vs IMPROVE(Kappa 0.771, P = 0.645), and Padua vs IMPROVE(Kappa 0.789, P = 0.812), whereas it did not present in the other pairs. The YI was improved to 0.304, 0.272, and 0.264 for the PERC(AUC 0.631[0.547-0.714], P = 0.006), Padua(AUC 0.613[0.527-0.700], P = 0.017), and IMPROVE(AUC 0.614[0.530-0.698], P = 0.016), with a revised cutoff of 5 or less, 6 or more, and 4 or more denoting the VTE-likely, respectively. CONCLUSIONS: For nonsurgical hospitalized patients with suspected VTE, the Geneva and Wells scores perform best, the PERC scores performs worst despite its significantly high sensitivity, whereas the others perform intermediately, albeit the absolute predictive power of all isolated scores are mediocre. The predictive power of the PERC, Padua, and IMPROVE scores are improved with revised cutoffs.

How large should the next study be? Predictive power and sample size requirements for replication studies.

We use information derived from over 40K trials in the Cochrane Collaboration database of systematic reviews (CDSR) to compute the replication probability, or predictive power of an experiment given its observed (two-sided) P$$ P $$ -value. We find that an exact replication of a marginally significant result with P=.05$$ P=.05 $$ has less than 30% chance of again reaching significance. Moreover, the replication of a result with P=.005$$ P=.005 $$ still has only 50% chance of significance. We also compute the probability that the direction (sign) of the estimated effect is correct, which is closely related to the type S error of Gelman and Tuerlinckx. We find that if an estimated effect has P=.05$$ P=.05 $$ , there is a 93% probability that its sign is correct. If P=.005$$ P=.005 $$ , then that probability is 99%. Finally, we compute the required sample size for a replication study to achieve some specified power conditional on the p$$ p $$ -value of the original study. We find that the replication of a result with P=.05$$ P=.05 $$ requires a sample size more than 16 times larger than the original study to achieve 80% power, while P=.005$$ P=.005 $$ requires at least 3.5 times larger sample size. These findings confirm that failure to replicate the statistical significance of a trial does not necessarily indicate that the original result was a fluke.

Conditional power and friends: The why and how of (un)planned, unblinded sample size recalculations in confirmatory trials.

Adapting the final sample size of a trial to the evidence accruing during the trial is a natural way to address planning uncertainty. Since the sample size is usually determined by an argument based on the power of the trial, an interim analysis raises the question of how the final sample size should be determined conditional on the accrued information. To this end, we first review and compare common approaches to estimating conditional power, which is often used in heuristic sample size recalculation rules. We then discuss the connection of heuristic sample size recalculation and optimal two-stage designs, demonstrating that the latter is the superior approach in a fully preplanned setting. Hence, unplanned design adaptations should only be conducted as reaction to trial-external new evidence, operational needs to violate the originally chosen design, or post hoc changes in the optimality criterion but not as a reaction to trial-internal data. We are able to show that commonly discussed sample size recalculation rules lead to paradoxical adaptations where an initially planned optimal design is not invariant under the adaptation rule even if the planning assumptions do not change. Finally, we propose two alternative ways of reacting to newly emerging trial-external evidence in ways that are consistent with the originally planned design to avoid such inconsistencies.

Conditional power in vaccine trials with seasonal variations.

Conditional power (CP) is widely used in clinical trial monitoring to quantify the evidence for futility stopping or sample size adaptation during the trial. When planning an interim analysis in vaccine trials for seasonal infectious diseases, CPs calculated under the hypothesized or currently estimated effect sizes may not truly reflect future data due to seasonal variations in disease incidence and/or vaccine efficacy (VE). Relying on these estimates alone could lead to erroneous decisions. Therefore, we carried out simulation studies to investigate the use of seven different choices for the drift parameter in computing CP or predictive power (PP) in end-of-season interim analysis. Our simulations showed that, when used to inform futility stopping, CP under the hypothesized effect and a weighted PP under a normal prior distribution appear to outperform others in terms of the overall type II error rate. All CPs and PPs considered in this study resulted in comparable powers and expected sample sizes when used to inform sample size adaptation. The performance of either CP or PP largely depends on the extent to which the chosen drift parameter or the prior distribution of the drift parameter matches the remainder of the trial. Weighted CP/PP tends to be less sensitive to settings where observed data and emerging data in future seasons differ substantially as they incorporate both current estimate and future variations. Therefore, weighted strategies deserve further exploration and perhaps increased usage in guiding trial operations because they are more robust to inaccuracies in prediction. In summary, for vaccine trials with seasonal variations, a decision on trial operations should be guided by a careful consideration of plausible CPs and PPs calculated under reasonable assumptions leveraging the data, prior hypotheses, and new evidence on clinical relevance.

A Probability-Based Models Ranking Approach: An Alternative Method of Machine-Learning Model Performance Assessment.

Performance measures are crucial in selecting the best machine learning model for a given problem. Estimating classical model performance measures by subsampling methods like bagging or cross-validation has several weaknesses. The most important ones are the inability to test the significance of the difference, and the lack of interpretability. Recently proposed Elo-based Predictive Power (EPP)-a meta-measure of machine learning model performance, is an attempt to address these weaknesses. However, the EPP is based on wrong assumptions, so its estimates may not be correct. This paper introduces the Probability-based Ranking Model Approach (PMRA), which is a modified EPP approach with a correction that makes its estimates more reliable. PMRA is based on the calculation of the probability that one model achieves a better result than another one, using the Mixed Effects Logistic Regression model. The empirical analysis was carried out on a real mortgage credits dataset. The analysis included a comparison of how the PMRA and state-of-the-art k-fold cross-validation ranked the 49 machine learning models, an example application of a novel method in hyperparameters tuning problem, and a comparison of PMRA and EPP indications. PMRA gives the opportunity to compare a newly developed algorithm to state-of-the-art algorithms based on statistical criteria. It is the solution to select the best hyperparameters configuration and to formulate criteria for the continuation of the hyperparameters space search.

Method for Data Quality Assessment of Synthetic Industrial Data.

Sometimes it is difficult, or even impossible, to acquire real data from sensors and machines that must be used in research. Such examples are the modern industrial platforms that frequently are reticent to share data. In such situations, the only option is to work with synthetic data obtained by simulation. Regarding simulated data, a limitation could consist in the fact that the data are not appropriate for research, based on poor quality or limited quantity. In such cases, the design of algorithms that are tested on that data does not give credible results. For avoiding such situations, we consider that mathematically grounded data-quality assessments should be designed according to the specific type of problem that must be solved. In this paper, we approach a multivariate type of prediction whose results finally can be used for binary classification. We propose the use of a mathematically grounded data-quality assessment, which includes, among other things, the analysis of predictive power of independent variables used for prediction. We present the assumptions that should be passed by the synthetic data. Different threshold values are established by a human assessor. In the case of research data, if all the assumptions pass, then we can consider that the data are appropriate for research and can be applied by even using other methods for solving the same type of problem. The applied method finally delivers a classification table on which can be applied any indicators of performed classification quality, such as sensitivity, specificity, accuracy, F1 score, area under curve (AUC), receiver operating characteristics (ROC), true skill statistics (TSS) and Kappa coefficient. These indicators' values offer the possibility of comparison of the results obtained by applying the considered method with results of any other method applied for solving the same type of problem. For evaluation and validation purposes, we performed an experimental case study on a novel synthetic dataset provided by the well-known UCI data repository.

A comparison of the four healthy days measures (HRQOL-4) with a single measure of self-rated general health in a population-based health survey in New York City.

BACKGROUND: Data on health-related quality of life (HRQOL) can be used to track health disparities, assess the impact of chronic diseases, and predict mortality. The Centers for Disease Control and Prevention's "Healthy Days Measures" (HRQOL-4) assesses four key domains: self-rated general health, physical health, mental health, and activity limitations. The domains are not easily combined to summarize overall HRQOL, and some evidence suggests that self-rated general health may be an adequate proxy indicator for overall HRQOL. This study compares self-rated general health as a solitary measure of HRQOL with two summary indices of the HRQOL-4 as a predictor of adverse health conditions in a representative sample of adult New York City residents. METHODS: The 2017 NYC Social Determinants of Health survey implemented by the New York City Department of Health and Mental Hygiene collected data from a representative sample of New Yorkers (n = 2335) via phone, mail, and web. We compared the information criteria and predictive power of self-rated general health with two alternative summary indices of the HRQOL-4 in predicting self-reported health conditions (hypertension, diabetes, obesity, non-specific psychological distress, and a summary indicator for at least one those four morbidities). RESULTS: Overall, 19.1% (95% CI: 16.9, 21.5) of respondents reported that they had fair or poor general health. Self-rated general health was significantly associated with days of poor physical health, poor mental health, and activity limitations (p < 0.001 for each). While the Akaike and Bayesian information criteria suggested that the summary indices of the HRQOL-4 produced marginally better models for predicting adverse health conditions, self-rated general health had slightly higher predictive power than did the summary indices in all models of physical health outcomes as measured by Tjur's pseudo-R2 and the area under the curve. CONCLUSION: We found very small differences between self-rated general health and the summary indices of the HRQOL-4 in predicting health conditions, suggesting self-rated general health is an appropriate proxy measure of overall HRQOL. Because it can be measured with a single question rather than four, it might be the most simple, efficient, and cost-effective method of summarizing HRQOL in large population-based surveys.

Fertility preferences and subsequent childbearing in Africa and Asia: A synthesis of evidence from longitudinal studies in 28 populations.

Survey data on fertility preferences have played a central but controversial role in fertility research and advocacy for family planning. We summarize evidence from longitudinal studies in 28 Asian and African populations on the relationship between preferences and subsequent childbearing. While we found no consistent association between women's desire to delay childbearing and subsequent fertility, the baseline desire of women to stop childbearing was a powerful predictor of subsequent fertility in all populations and increased in strength as overall contraceptive use in the study populations rose. Partners' desire also exercised some influence but was of modest importance in most populations. However, the correspondence between desire to stop and behaviour was found to be far from perfect. Weak implementation of preferences by contraception is likely to be the major cause of this preference-behaviour discrepancy. Uncertainty and instability in preferences may also contribute to the discrepancy, particularly in sub-Saharan Africa.

Utilizing Bayesian predictive power in clinical trial design.

The Bayesian paradigm provides an ideal platform to update uncertainties and carry them over into the future in the presence of data. Bayesian predictive power (BPP) reflects our belief in the eventual success of a clinical trial to meet its goals. In this paper we derive mathematical expressions for the most common types of outcomes, to make the BPP accessible to practitioners, facilitate fast computations in adaptive trial design simulations that use interim futility monitoring, and propose an organized BPP-based phase II-to-phase III design framework.

Unique predictive power of other-rated personality: An 18-year longitudinal study.

OBJECTIVE: What is gained by having others report on one's personality? Research on adult samples has suggested that informant reports are especially informative regarding traits that are highly visible and evaluative (i.e., socially desirable/undesirable instead of neutral), such as Openness, Conscientiousness, and Agreeableness. This 18-year longitudinal study aims to demonstrate the unique predictive power of other-rated personality in adolescence, using life outcomes and personality in adulthood as trait criteria. METHOD: We examined the unique predictive power of self- and other-rated Big Five personality traits at age 12 and 17 on self-rated life outcomes and personality at age 29 (e.g., educational achievement, work income, depression, moral transgressions, and relationship satisfaction). Participants were 186 German adolescents (53% boys), their parents and friends at age 12, and their mothers and fathers at age 17. RESULTS: Other-ratings showed unique predictive power beyond self-ratings for all Big Five traits, with the most consistent results for Openness, Conscientiousness, and Agreeableness. CONCLUSIONS: Results demonstrate the added value of including other-reports on adolescent personality when predicting future life outcomes and personality, especially for highly visible and evaluative traits. The present study sheds light on the predictive power of self- versus other-rated personality and personality-outcome associations.

Fas-ligand and interleukin-6 in the cerebrospinal fluid are early predictors of hypoxic-ischemic encephalopathy and long-term outcomes after birth asphyxia in term infants.

BACKGROUND: Cerebral ischemia generates neuroinflammation that can induce neural cell death. This cohort study assessed whether Fas-ligand (FasL) and interleukin (IL)-6 levels in the cerebrospinal fluid (CSF) after hypoxic-ischemic encephalopathy (HIE) can serve as biomarkers of hypoxic brain injury in neonates. METHODS: Term infants (> 37-week gestational age) who were admitted to the neonatal intensive care unit of Karolinska University Hospital in years 2002 to 2004 with perinatal asphyxia were enrolled prospectively. Control infants without brain pathology underwent lumbar puncture for suspected infection. FasL and IL-6 levels were measured in the CSF, by enzyme-linked immunosorbent assays. All patients underwent neurological assessment at 18 months. HIE was classified as mild, moderate, or severe (HIE I-III). Adverse neurological outcome at 18 months was defined as a mental developmental index < 85, deafness, blindness, cerebral palsy, or seizure disorder. RESULTS: Of the 44 HIE patients, 14, 16, and 14 had HIE-I, HIE-II, and HIE-III, respectively. HIE-II and HIE-III patients had higher FasL and IL-6 levels than HIE-I patients and the 20 controls (all p < 0.0001). Patients with adverse outcomes had higher FasL and IL-6 levels than patients with normal outcomes and controls (both p < 0.0001). On receiver-operator curve analyses, FasL and IL-6 (alone and together) were highly predictive of HIE grade and outcome (areas under the curve range 0.86-0.94) and showed high sensitivity (66.7-100%). These biomarkers performed better than cord blood pH (areas under the curve: HIE grade = 0.80, adverse outcomes = 0.86). CONCLUSION: CSF biomarkers FasL and IL-6 predicted severity of encephalopathy and long-term outcomes in post-asphyxiated infants better than a standard biomarker.

Conditional power and predictive power based on right censored data with supplementary auxiliary information.

Conditional power and predictive power provide estimates of the probability of success at the end of the trial based on the information from the interim analysis. The observed value of the time to event endpoint at the interim analysis could be biased for the true treatment effect due to early censoring, leading to a biased estimate of conditional power and predictive power. In such cases, the estimates and inference for this right censored primary endpoint are enhanced by incorporating a fully observed auxiliary variable. We assume a bivariate normal distribution of the transformed primary variable and a correlated auxiliary variable. Simulation studies are conducted that not only shows enhanced conditional power and predictive power but also can provide the framework for a more efficient futility interim analysis in terms of an improved accuracy in estimator, a smaller inflation in type II error and an optimal timing for such analysis. We also illustrated the new approach by a real clinical trial example.