Explore the vast range of titles available.

N-of-1 trials allow inference between two treatments given to a single individual. Most often, clinical investigators analyze an individual's N-of-1 trial data with usual t-tests or simple nonparametric methods. These simple methods do not account for serial correlation in repeated observations coming from the individual. Existing methods accounting for serial correlation require simulation, multiple N-of-1 trials, or both. Here, we develop t-tests that account for serial correlation in a single individual. The development includes effect size and precision calculations, both of which are useful for study planning. We then use Monte Carlo simulation to evaluate statistical properties of these serial t-tests, namely, Type I and II errors, and confidence interval widths, and compare these statistical properties to those of analogous usual t-test. The serial t-tests clearly outperform the usual t-tests commonly used in reporting N-of-1 results. Examples from N-of-1 clinical trials in fibromyalgia patients and from a behavioral health setting exhibit how accounting for serial correlation can change inferences. These t-tests are easily implemented and more appropriate than simple methods commonly used; however, caution is needed when analyzing only a few observations.

N-of-1 trials allow inference between two treatments given to a single individual. Most often, clinical investigators analyze an individual's N-of-1 trial data with usual t-tests or simple nonparametric methods. These simple methods do not account for serial correlation in repeated observations coming from the individual. Existing methods accounting for serial correlation require simulation, multiple N-of-1 trials, or both. Here, we develop t-tests that account for serial correlation in a single individual. The development includes effect size and precision calculations, both of which are useful for study planning. We then evaluate and compare their Type I and II errors and interval estimators to those of usual t-tests analogues via Monte Carlo simulation. The serial t-tests clearly outperform the usual t-tests commonly used in reporting N-of-1 results. Examples from N-of-1 clinical trials in fibromyalgia patients and from a behavioral health setting exhibit how accounting for serial correlation can change inferences. These t-tests are easily implemented and more appropriate than simple methods commonly used; however, caution is needed when analyzing only a few observations. Keywords: Autocorrelation; Cross-over studies; Repeated measures analysis; Single-case experimental design; Time-series

Generative models of graphs are well-known, but many existing models are limited in scalability and expressivity. We present a novel sequential graphical variational autoencoder operating directly on graphical representations of data. In our model, the encoding and decoding of a graph as is framed as a sequential deconstruction and construction process, respectively, enabling the the learning of a latent space. Experiments on a cycle dataset show promise, but highlight the need for a relaxation of the distribution over node permutations.

We present Chirpy Cardinal, an open-domain social chatbot. Aiming to be both informative and conversational, our bot chats with users in an authentic, emotionally intelligent way. By integrating controlled neural generation with scaffolded, hand-written dialogue, we let both the user and bot take turns driving the conversation, producing an engaging and socially fluent experience. Deployed in the fourth iteration of the Alexa Prize Socialbot Grand Challenge, Chirpy Cardinal handled thousands of conversations per day, placing second out of nine bots with an average user rating of 3.58/5.

Language models demonstrate both quantitative improvement and new qualitative capabilities with increasing scale. Despite their potentially transformative impact, these new capabilities are as yet poorly characterized. In order to inform future research, prepare for disruptive new model capabilities, and ameliorate socially harmful effects, it is vital that we understand the present and near-future capabilities and limitations of language models. To address this challenge, we introduce the Beyond the Imitation Game benchmark (BIG-bench). BIG-bench currently consists of 204 tasks, contributed by 450 authors across 132 institutions. Task topics are diverse, drawing problems from linguistics, childhood development, math, common-sense reasoning, biology, physics, social bias, software development, and beyond. BIG-bench focuses on tasks that are believed to be beyond the capabilities of current language models. We evaluate the behavior of OpenAI's GPT models, Google-internal dense transformer architectures, and Switch-style sparse transformers on BIG-bench, across model sizes spanning millions to hundreds of billions of parameters. In addition, a team of human expert raters performed all tasks in order to provide a strong baseline. Findings include: model performance and calibration both improve with scale, but are poor in absolute terms (and when compared with rater performance); performance is remarkably similar across model classes, though with benefits from sparsity; tasks that improve gradually and predictably commonly involve a large knowledge or memorization component, whereas tasks that exhibit \"breakthrough\" behavior at a critical scale often involve multiple steps or components, or brittle metrics; social bias typically increases with scale in settings with ambiguous context, but this can be improved with prompting.

Germ granules are membrane-less organelles essential for small RNA biogenesis and germline development. Among the conserved properties of germ granules is their association with the nuclear membrane. Recent studies demonstrated that LOTUS domain proteins, EGGD-1 and EGGD-2 (also known as MIP-1 and MIP-2 respectively), promote the formation of perinuclear germ granules in C. elegans . This finding presents a unique opportunity to evaluate the significance of perinuclear localization of germ granules. Here we show that loss of eggd-1 causes the coalescence of germ granules and formation of abnormal cytoplasmic aggregates. Impairment of perinuclear granules affects certain germline classes of small RNAs including Piwi-interacting RNAs. Transcriptome profiling reveals overexpression of spermatogenic and cuticle-related genes in eggd-1 hermaphrodites. We further demonstrate that disruption of germ granules activates HLH-30-mediated transcriptional program in somatic tissues. Collectively, our findings underscore the essential role of EGGD-1 in germ granule organization and reveal an unexpected germ granule-to-soma communication. Germ granules are membraneless organelles that act as organizing centers for small RNA biogenesis during germline development. Here they show the LOTUS domain protein eggd-1 organizes germ granules, and reveal a germ line-to-soma communication pathway activated upon perturbation of germ granules.

Metastatic breast cancer is usually diagnosed after becoming symptomatic, at which point it is rarely curable. Cell‐free circulating tumor DNA (ctDNA) contains tumor‐specific chromosomal rearrangements that may be interrogated in blood plasma. We evaluated serial monitoring of ctDNA for earlier detection of metastasis in a retrospective study of 20 patients diagnosed with primary breast cancer and long follow‐up. Using an approach combining low‐coverage whole‐genome sequencing of primary tumors and quantification of tumor‐specific rearrangements in plasma by droplet digital PCR, we identify for the first time that ctDNA monitoring is highly accurate for postsurgical discrimination between patients with (93%) and without (100%) eventual clinically detected recurrence. ctDNA‐based detection preceded clinical detection of metastasis in 86% of patients with an average lead time of 11 months (range 0–37 months), whereas patients with long‐term disease‐free survival had undetectable ctDNA postoperatively. ctDNA quantity was predictive of poor survival. These findings establish the rationale for larger validation studies in early breast cancer to evaluate ctDNA as a monitoring tool for early metastasis detection, therapy modification, and to aid in avoidance of overtreatment. Synopsis Serial measurement of circulating tumor DNA (ctDNA) is shown to be a robust and accurate occult metastatic disease biomarker in patients diagnosed with primary breast cancer. Measured ctDNA levels are a quantitative risk factor for poor outcomes. A combination of low‐coverage whole‐genome sequencing with personalized droplet digital PCR, analytical methods, and a bioinformatics pipeline was developed for quantification of ctDNA in blood plasma samples collected during clinical follow‐up of patients with primary (non‐metastatic) breast cancer. ctDNA analysis can discriminate patients with eventual metastasis from those with long‐term disease‐free survival with 93% sensitivity and 100% specificity (ROC area 0.98, P  = 0.001). ctDNA‐based detection of occult metastatic disease preceded clinical detection for 86% of patients by an average 11 months and in some cases by 3 years. No ctDNA could be detected at any time‐point after surgery for patients with long‐term disease‐free survival. The level of ctDNA was a quantitative risk factor for clinical metastatic disease (logistic regression odds ratio 2.1 for each doubling of ctDNA levels, P  = 0.02) and death (odds ratio 1.3 per ctDNA doubling, P  = 0.04). Graphical Abstract Serial measurement of circulating tumor DNA (ctDNA) is shown to be a robust and accurate occult metastatic disease biomarker in patients diagnosed with primary breast cancer. Measured ctDNA levels are a quantitative risk factor for poor outcomes.

Premature infants in neonatal intensive care units (NICUs) have underdeveloped immune systems, making them susceptible to adverse health consequences from air pollutant exposure. Little is known about the sources of indoor airborne particles that contribute to the exposure of premature infants in the NICU environment. In this study, we monitored the spatial and temporal variations of airborne particulate matter concentrations along with other indoor environmental parameters and human occupancy. The experiments were conducted over one year in a private-style NICU. The NICU was served by a central heating, ventilation and air-conditioning (HVAC) system equipped with an economizer and a high-efficiency particle filtration system. The following parameters were measured continuously during weekdays with 1-min resolution: particles larger than 0.3 μm resolved into 6 size groups, CO2 level, dry-bulb temperature and relative humidity, and presence or absence of occupants. Altogether, over sixteen periods of a few weeks each, measurements were conducted in rooms occupied with premature infants. In parallel, a second monitoring station was operated in a nearby hallway or at the local nurses' station. The monitoring data suggest a strong link between indoor particle concentrations and human occupancy. Detected particle peaks from occupancy were clearly discernible among larger particles and imperceptible for submicron (0.3-1 μm) particles. The mean indoor particle mass concentrations averaged across the size range 0.3-10 μm during occupied periods was 1.9 μg/m(3), approximately 2.5 times the concentration during unoccupied periods (0.8 μg/m(3)). Contributions of within-room emissions to total PM10 mass in the baby rooms averaged 37-81%. Near-room indoor emissions and outdoor sources contributed 18-59% and 1-5%, respectively. Airborne particle levels in the size range 1-10 μm showed strong dependence on human activities, indicating the importance of indoor-generated particles for infant's exposure to airborne particulate matter in the NICU.

In this qualitative study, we research what constitutes the relationships between conceptions of quality and value associated with higher education as experienced by prospective employers of business graduates. Quality and value in higher education are often linked though the relationship is unclear. Employers are an important and under-researched stakeholder group on the demand side of higher education. Data are generated by interviewing prospective employers of business graduates from a UK university. Interviews are analysed using a phenomenographic method to determine the qualitatively different ways in which actors make sense of the relationships between quality and value. Understanding prospective employers' conceptions of the relationships is important given the competitive pressures on universities and businesses. The research reinforces the experiential and idiosyncratic relationships between quality and value in higher education. Three conceptions of what constitutes quality and value in higher education are discussed: (a) quality is an antecedent of value; (b) quality is simple while value is complex; and (c) quality is internal to HE while value is created in the customer domain. The research outcomes provide important insights for researchers and practitioners through clearer understanding of how quality and value are related for this important stakeholder group. (HRK / Abstract übernommen).

Inhibition of Janus kinase (JAK) family enzymes is a popular strategy for treating inflammatory and autoimmune skin diseases. In the clinic, small molecule JAK inhibitors show distinct efficacy and safety profiles, likely reflecting variable selectivity for JAK subtypes. Absolute JAK subtype selectivity has not yet been achieved. Here, we rationally design small interfering RNAs (siRNAs) that offer sequence-specific gene silencing of JAK1, narrowing the spectrum of action on JAK-dependent cytokine signaling to maintain efficacy and improve safety. Our fully chemically modified siRNA supports efficient silencing of JAK1 expression in human skin explant and modulation of JAK1-dependent inflammatory signaling. A single injection into mouse skin enables five weeks of duration of effect. In a mouse model of vitiligo, local administration of the JAK1 siRNA significantly reduces skin infiltration of autoreactive CD8 + T cells and prevents epidermal depigmentation. This work establishes a path toward siRNA treatments as a new class of therapeutic modality for inflammatory and autoimmune skin diseases. Therapeutic modulation of Janus kinase family enzymes is an established approach for inflammatory and autoimmune skin diseases. Here the authors rationally design small interfering RNAs to enable single Janus kinase targeting and test this new therapeutic approach in a skin disease model for maintaining efficacy and improving selectivity.