To ensure undisrupted business, large Internet companies need to closely monitor various KPIs (e.g., Page Views, number of online users, and number of orders) of its Web applications, to accurately detect anomalies and trigger timely troubleshooting/mitigation. However, anomaly detection for these seasonal KPIs with various patterns and data quality has been a great challenge, especially without labels. In this paper, we proposed Donut, an unsupervised anomaly detection algorithm based on VAE. Thanks to a few of our key techniques, Donut greatly outperforms a state-of-arts supervised ensemble approach and a baseline VAE approach, and its best F-scores range from 0.75 to 0.9 for the studied KPIs from a top global Internet company. We come up with a novel KDE interpretation of reconstruction for Donut, making it the first VAE-based anomaly detection algorithm with solid theoretical explanation.

Objective: To explore metabolic disturbances in nonobese women with polycystic ovary syndrome (PCOS) compared with nonobese healthy controls. Design: Systematic review and meta-analysis. Setting: Not applicable. Patient(s): Nonobese women with PCOS and nonobese healthy controls. Intervention(s): None. Main Outcome Measure(s): Prevalence of metabolic disturbances including hyperinsulinemia, insulin resistance (IR), impaired fasting glucose (IFG), impaired glucose intolerance (IGT), prediabetes, dyslipidemia, hypercholesterolemia, hypertriglyceridemia, and low high-density lipoprotein (low-HDL), as well as other metabolic outcomes such as type 2 diabetes mellitus (T2DM), hypertension, metabolic syndrome (Mets), myocardial infarction, stroke, cerebrovascular accident, arterial occlusive disease, and coronary heart disease. Result(s): Compared to nonobese controls, nonobese women with PCOS showed a higher prevalence of hyperinsulinemia (odds ratio [

α-Amyrin is a plant-derived pentacyclic triterpenoid, with a lot of important physiological and pharmacological activities. The formation of α-amyrin from (3 S)-2,3-oxidosqualene is catalyzed by α-amyrin synthase (α-AS), a member of the oxidosqualene cyclase (OSC) protein family. However, α-amyrin is not yet commercially developed due to its extremely low productivity in plants. The engineered Saccharomyces cerevisiae with efficient α-amyrin production pathway could be used as an alternative and sustainable solution to produce α-amyrin from renewable raw materials. To efficiently improve α-amyrin production in S. cerevisiae, we identified two α-ASs, EjAS and MdOSC1 from Eriobotrya japonica and Malus × domestica, respectively, through strict bioinformatics screening criteria and phylogenetic analysis. The specific activities of purified EjAS and MdOSC1 were 0.0032 and 0.0293 μmol/min/mg, respectively. EjAS produced α-amyrin and β-amyrin at a ratio of 17:3, MdOSC1 produced α-amyrin, β-amyrin and lupeol at a ratio of 86:13:1, indicating MdOSC1 had significantly higher specific activity and higher ratio of α-amyrin than EjAS. Furthermore, MdOSC1 was introduced into S. cerevisiae combining with the increased supply of (3 S)-2,3-oxidosqualene to achieve the encouraging α-amyrin production, and the titer of α-amyrin achieved 11.97 ± 0.61 mg/L, 5.8 folds of the maximum production reported.

The Multiple-Network Poroelastic Theory (MPET) is a numerical model to characterize the transport of multiple fluid networks in the brain, which overcomes the problem of conducting separate analyses on individual fluid compartments and losing the interactions between tissue and fluids, in addition to the interaction between the different fluids themselves. In this paper, the blood perfusion results from MPET modeling are partially validated using cerebral blood flow (CBF) data obtained from arterial spin labeling (ASL) magnetic resonance imaging (MRI), which uses arterial blood water as an endogenous tracer to measure CBF. Two subjects—one healthy control and one patient with unilateral middle cerebral artery (MCA) stenosis are included in the validation test. The comparison shows several similarities between CBF data from ASL and blood perfusion results from MPET modeling, such as higher blood perfusion in the gray matter than in the white matter, higher perfusion in the periventricular region for both the healthy control and the patient, and asymmetric distribution of blood perfusion for the patient. Although the partial validation is mainly conducted in a qualitative way, it is one important step toward the full validation of the MPET model, which has the potential to be used as a testing bed for hypotheses and new theories in neuroscience research.

No abstract