Identifying Beneficial Learning Behaviors from Large-Scale Interaction Data
Ontology highlight
ABSTRACT:
SUBMITTER: Bittencourt I
PROVIDER: S-EPMC7334710 | biostudies-literature | 2020 Jun
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Although several hundred regions of the human genome harbor signals of positive natural selection, few of the relevant adaptive traits and variants have been elucidated. Using full-genome sequence variation from the 1000 Genomes (1000G) Project and the composite of multiple signals (CMS) test, we investigated 412 candidate signals and leveraged functional annotation, protein structure modeling, epigenetics, and association studies to identify and extensively annotate candidate causal variants. The resulting catalog provides a tractable list for experimental follow-up; it includes 35 high-scoring nonsynonymous variants, 59 variants associated with expression levels of a nearby coding gene or lincRNA, and numerous variants associated with susceptibility to infectious disease and other phenotypes. We experimentally characterized one candidate nonsynonymous variant in Toll-like receptor 5 (TLR5) and show that it leads to altered NF-?B signaling in response to bacterial flagellin. PAPERFLICK:
| S-EPMC3674781 | biostudies-literature
Project description:Understanding the complex dynamics of the atmosphere is of paramount interest due to its impact in the entire climate system and in human society. Here we focus on identifying, from data, the geographical regions which have similar atmospheric properties. We study surface air temperature (SAT) time series with monthly resolution, recorded at a regular grid covering the Earth surface. We consider two datasets: NCEP CDAS1 and ERA Interim reanalysis. We show that two surprisingly simple measures are able to extract meaningful information: i) the distance between the lagged SAT and the incoming solar radiation and ii) the Shannon entropy of SAT and SAT anomalies. The distance uncovers well-defined spatial patterns formed by regions with similar SAT response to solar forcing while the entropy uncovers regions with similar degree of SAT unpredictability. The entropy analysis also allows identifying regions in which SAT has extreme values. Importantly, we uncover differences between the two datasets which are due to the presence of extreme values in one dataset but not in the other. Our results indicate that the distance and entropy measures can be valuable tools for the study of other climatological variables, for anomaly detection and for performing model inter-comparisons.
| S-EPMC5372476 | biostudies-literature
Project description:Recently, approaches have been developed for high-throughput identification of synthetic sick/lethal gene pairs. However, these are only a specific example of the broader phenomenon of epistasis, wherein the presence of one mutation modulates the phenotype of another. We present analysis techniques for generating high-confidence quantitative epistasis scores from measurements made using synthetic genetic array and epistatic miniarray profile (E-MAP) technology, as well as several tools for higher-level analysis of the resulting data that are greatly enhanced by the quantitative score and detection of alleviating interactions.
| S-EPMC1779568 | biostudies-literature
Project description:BackgroundDigital learning environments have become very common in the training of medical professionals, and students often use such platforms for exam preparation. Multiple choice questions (MCQs) are a common format in medical exams and are used by students to prepare for said exams.ObjectiveWe aimed to examine whether particular learning activities contributed more strongly than others to users' exam performance.MethodsWe analyzed data from users of an online platform that provides learning materials for medical students in preparation for their final exams. We analyzed whether the number of learning cards viewed and the number of MCQs taken were positively related to learning outcomes. We also examined whether viewing learning cards or answering MCQs was more effective. Finally, we tested whether taking individual notes predicted learning outcomes, and whether taking notes had an effect after controlling for the effects of learning cards and MCQs. Our analyses from the online platform Amboss are based on user activity data, which supplied the number of learning cards studied and test questions answered. We also included the number of notes from each of those 23,633 users who had studied at least 200 learning cards and had answered at least 1000 test exam questions in the 180 days before their state exam. The activity data for this analysis was collected retrospectively, using Amboss archival usage data from April 2014 to April 2017. Learning outcomes were measured using the final state exam scores that were calculated by using the answers voluntarily entered by the participants.ResultsWe found correlations between the number of cards studied (r=.22; P<.001) and the number of test questions that had been answered (r=.23; P<.001) with the percentage of correct answers in the learners' medical exams. The number of test questions answered still yielded a significant effect, even after controlling for the number of learning cards studied using a hierarchical regression analysis (?=.14; P<.001; ?R2=.017; P<.001). We found a negative interaction between the number of learning cards and MCQs, indicating that users with high scores for learning cards and MCQs had the highest exam scores. Those 8040 participants who had taken at least one note had a higher percentage of correct answers (80.94%; SD=7.44) than those who had not taken any notes (78.73%; SD=7.80; t23631=20.95; P<.001). In a stepwise regression, the number of notes the participants had taken predicted the percentage of correct answers over and above the effect of the number of learning cards studied and of the number of test questions entered in step one (?=.06; P<.001; ?R2=.004; P<.001).ConclusionsThese results show that online learning platforms are particularly helpful whenever learners engage in active elaboration in learning material, such as by answering MCQs or taking notes.
| S-EPMC6724501 | biostudies-literature
Project description:BACKGROUND: Single nucleotide polymorphism (SNP) based association studies aim at identifying SNPs associated with phenotypes, for example, complex diseases. The associated SNPs may influence the disease risk individually (main effects) or behave jointly (epistatic interactions). For the analysis of high throughput data, the main difficulty is that the number of SNPs far exceeds the number of samples. This difficulty is amplified when identifying interactions. RESULTS: In this paper, we propose an Adaptive Group Lasso (AGL) model for large-scale association studies. Our model enables us to analyze SNPs and their interactions simultaneously. We achieve this by introducing a sparsity constraint in our model based on the fact that only a small fraction of SNPs is disease-associated. In order to reduce the number of false positive findings, we develop an adaptive reweighting scheme to enhance sparsity. In addition, our method treats SNPs and their interactions as factors, and identifies them in a grouped manner. Thus, it is flexible to analyze various disease models, especially for interaction detection. However, due to the intensive computation when millions of interaction terms needs to be searched in the model fitting, our method needs to combined with some filtering methods when applied to genome-wide data for detecting interactions. CONCLUSION: By using a wide range of simulated datasets and a real dataset from WTCCC, we demonstrate the advantages of our method.
| S-EPMC3203332 | biostudies-literature
Project description:MotivationMonoclonal antibody therapeutics are often produced from non-human sources (typically murine), and can therefore generate immunogenic responses in humans. Humanization procedures aim to produce antibody therapeutics that do not elicit an immune response and are safe for human use, without impacting efficacy. Humanization is normally carried out in a largely trial-and-error experimental process. We have built machine learning classifiers that can discriminate between human and non-human antibody variable domain sequences using the large amount of repertoire data now available.ResultsOur classifiers consistently outperform the current best-in-class model for distinguishing human from murine sequences, and our output scores exhibit a negative relationship with the experimental immunogenicity of existing antibody therapeutics. We used our classifiers to develop a novel, computational humanization tool, Hu-mAb, that suggests mutations to an input sequence to reduce its immunogenicity. For a set of therapeutic antibodies with known precursor sequences, the mutations suggested by Hu-mAb show significant overlap with those deduced experimentally. Hu-mAb is therefore an effective replacement for trial-and-error humanization experiments, producing similar results in a fraction of the time.AvailabilityHu-mAb (humanness scoring and humanization) is freely available to use at opig.stats.ox.ac.uk/webapps/humab.Supplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC8760955 | biostudies-literature
Project description:The brain performs various cognitive functions by learning the spatiotemporal salient features of the environment. This learning requires unsupervised segmentation of hierarchically organized spike sequences, but the underlying neural mechanism is only poorly understood. Here, we show that a recurrent gated network of neurons with dendrites can efficiently solve difficult segmentation tasks. In this model, multiplicative recurrent connections learn a context-dependent gating of dendro-somatic information transfers to minimize error in the prediction of somatic responses by the dendrites. Consequently, these connections filter the redundant input features represented by the dendrites but unnecessary in the given context. The model was tested on both synthetic and real neural data. In particular, the model was successful for segmenting multiple cell assemblies repeating in large-scale calcium imaging data containing thousands of cortical neurons. Our results suggest that recurrent gating of dendro-somatic signal transfers is crucial for cortical learning of context-dependent segmentation tasks.
| S-EPMC9249189 | biostudies-literature
Project description:Sedentary lifestyle is an escalating epidemic. Little is known about whether or how social media can be used to design a cost-effective solution for sedentary lifestyle. In this article we describe the data from a randomized controlled trial (RCT) that evaluated two prominent strategies for conducting exercise interventions using elements of social media: motivational media campaigns and online peer networks. The data file includes 217 participants' basic demographic information, number of exercise class enrollments over 13 weeks, and self-reported number of days for exercise activities in the previous 7 days at baseline. Among the 217, 164 also have data on self-reported number of days for exercise activities at the post-program. Data are supplied with this article. The interpretation of these data can be found in the research article published by the authors in Preventive Medicine Reports in 2015 [1].
| S-EPMC4610960 | biostudies-literature
Project description:BackgroundIn recent years the field of movement ecology has been revolutionized by our ability to collect high-accuracy, fine scale telemetry data from individual animals and groups. This growth in our data collection capacity has led to the development of statistical techniques that integrate telemetry data with random walk models to infer key parameters of the movement dynamics. While much progress has been made in the use of these models, several challenges remain. Notably robust and scalable methods are required for quantifying parameter uncertainty, coping with intermittent location fixes, and analysing the very large volumes of data being generated.MethodsIn this work we implement a novel approach to movement modelling through the use of multilevel Gaussian processes. The hierarchical structure of the method enables the inference of continuous latent behavioural states underlying movement processes. For efficient inference on large data sets, we approximate the full likelihood using trajectory segmentation and sample from posterior distributions using gradient-based Markov chain Monte Carlo methods.ResultsWhile formally equivalent to many continuous-time movement models, our Gaussian process approach provides flexible, powerful models that can detect multiscale patterns and trends in movement trajectory data. We illustrate a further advantage to our approach in that inference can be performed using highly efficient, GPU-accelerated machine learning libraries.ConclusionsMultilevel Gaussian process models offer efficient inference for large-volume movement data sets, along with the fitting of complex flexible models. Applications of this approach include inferring the mean location of a migration route and quantifying significant changes, detecting diurnal activity patterns, or identifying the onset of directed persistent movements.
| S-EPMC7893961 | biostudies-literature
Project description:Nationwide population-based cohort provides a new opportunity to build an automated risk prediction model based on individuals' history of health and healthcare beyond existing risk prediction models. We tested the possibility of machine learning models to predict future incidence of Alzheimer's disease (AD) using large-scale administrative health data. From the Korean National Health Insurance Service database between 2002 and 2010, we obtained de-identified health data in elders above 65 years (N = 40,736) containing 4,894 unique clinical features including ICD-10 codes, medication codes, laboratory values, history of personal and family illness and socio-demographics. To define incident AD we considered two operational definitions: "definite AD" with diagnostic codes and dementia medication (n = 614) and "probable AD" with only diagnosis (n = 2026). We trained and validated random forest, support vector machine and logistic regression to predict incident AD in 1, 2, 3, and 4 subsequent years. For predicting future incidence of AD in balanced samples (bootstrapping), the machine learning models showed reasonable performance in 1-year prediction with AUC of 0.775 and 0.759, based on "definite AD" and "probable AD" outcomes, respectively; in 2-year, 0.730 and 0.693; in 3-year, 0.677 and 0.644; in 4-year, 0.725 and 0.683. The results were similar when the entire (unbalanced) samples were used. Important clinical features selected in logistic regression included hemoglobin level, age and urine protein level. This study may shed a light on the utility of the data-driven machine learning model based on large-scale administrative health data in AD risk prediction, which may enable better selection of individuals at risk for AD in clinical trials or early detection in clinical settings.
| S-EPMC7099065 | biostudies-literature