The Problem with Big Data: Operating on Smaller Datasets to Bridge the Implementation Gap.
Ontology highlight
ABSTRACT:
SUBMITTER: Mann RP
PROVIDER: S-EPMC5130981 | biostudies-literature | 2016
REPOSITORIES: biostudies-literature
Publications
The Problem with Big Data: Operating on Smaller Datasets to Bridge the Implementation Gap.
Mann Richard P RP Mushtaq Faisal F White Alan D AD Mata-Cervantes Gabriel G Pike Tom T Coker Dalton D Murdoch Stuart S Hiles Tim T Smith Clare C Berridge David D Hinchliffe Suzanne S Hall Geoff G Smye Stephen S Wilkie Richard M RM Lodge J Peter A JP Mon-Williams Mark M
Frontiers in public health 20161201
Big datasets have the potential to revolutionize public health. However, there is a mismatch between the political and scientific optimism surrounding big data and the public's perception of its benefit. We suggest a systematic and concerted emphasis on developing models derived from smaller datasets to illustrate to the public how big data can produce tangible benefits in the long term. In order to highlight the immediate value of a small data approach, we produced a proof-of-concept model pred ...[more]
Similar Datasets
Project description:Despite the important discoveries reported by genome-wide association (GWA) studies, for most traits and diseases the prediction R-squared (R-sq.) achieved with genetic scores remains considerably lower than the trait heritability. Modern biobanks will soon deliver unprecedentedly large biomedical data sets: Will the advent of big data close the gap between the trait heritability and the proportion of variance that can be explained by a genomic predictor? We addressed this question using Bayesian methods and a data analysis approach that produces a surface response relating prediction R-sq. with sample size and model complexity (e.g., number of SNPs). We applied the methodology to data from the interim release of the UK Biobank. Focusing on human height as a model trait and using 80,000 records for model training, we achieved a prediction R-sq. in testing (n = 22,221) of 0.24 (95% C.I.: 0.23-0.25). Our estimates show that prediction R-sq. increases with sample size, reaching an estimated plateau at values that ranged from 0.1 to 0.37 for models using 500 and 50,000 (GWA-selected) SNPs, respectively. Soon much larger data sets will become available. Using the estimated surface response, we forecast that larger sample sizes will lead to further improvements in prediction R-sq. We conclude that big data will lead to a substantial reduction of the gap between trait heritability and the proportion of interindividual differences that can be explained with a genomic predictor. However, even with the power of big data, for complex traits we anticipate that the gap between prediction R-sq. and trait heritability will not be fully closed.
| S-EPMC5676235 | biostudies-literature
Project description:In many experiments and especially in translational and preclinical research, sample sizes are (very) small. In addition, data designs are often high dimensional, i.e. more dependent than independent replications of the trial are observed. The present paper discusses the applicability of max t-test-type statistics (multiple contrast tests) in high-dimensional designs (repeated measures or multivariate) with small sample sizes. A randomization-based approach is developed to approximate the distribution of the maximum statistic. Extensive simulation studies confirm that the new method is particularly suitable for analyzing data sets with small sample sizes. A real data set illustrates the application of the methods.
| S-EPMC8008424 | biostudies-literature
Project description:Three synthetic datasets - of observation size 15,000, 155,000 and 1,555,000 participants, respectively - were created by simulating eleven cardiac and anthropometric variables from nine collection ages of the ALSAPC birth cohort study. The synthetic datasets retain similar data properties to the ALSPAC study data they are simulated from (co-variance matrices, as well as the mean and variance values of the variables) without including the original data itself or disclosing participant information. In this instance, the three synthetic datasets have been utilised in an academia-industry collaboration to build a prototype virtual reality data analysis software, but they could have a broader use in method and software development projects where sensitive data cannot be freely shared.
| S-EPMC5605951 | biostudies-literature
Project description:The genomic inter-individual heterogeneity remains a significant challenge for both clinical decision-making and the design of clinical trials. Although next-generation sequencing (NGS) is increasingly implemented in drug development and clinical trials, translation of the obtained genomic information into actionable clinical advice lags behind. Major reasons are the paucity of sufficiently powered trials that can quantify the added value of pharmacogenetic testing, and the considerable pharmacogenetic complexity with millions of rare variants with unclear functional consequences. The resulting uncertainty is reflected in inconsistencies of pharmacogenomic drug labels in Europe and the United States. In this review, we discuss how the knowledge gap for bridging pharmacogenomics into the clinics can be reduced. First, emerging methods that allow the high-throughput experimental characterization of pharmacogenomic variants combined with novel computational tools hold promise to improve the accuracy of drug response predictions. Second, tapping of large biobanks of therapeutic drug monitoring data allows to conduct high-powered retrospective studies that can validate the clinical importance of genetic variants, which are currently incompletely characterized. Combined, we are confident that these methods will improve the accuracy of drug response predictions and will narrow the gap between variant identification and its utilization for clinical decision-support.
| S-EPMC7057970 | biostudies-literature
Project description:BackgroundObesity research at a population level is multifaceted and complex. This has been characterised in the UK by the Foresight obesity systems map, identifying over 100 variables, across seven domain areas which are thought to influence energy balance, and subsequent obesity. Availability of data to consider the whole obesity system is traditionally lacking. However, in an era of big data, new possibilities are emerging. Understanding what data are available can be the first challenge, followed by an inconsistency in data reporting to enable adequate use in the obesity context. In this study we map data sources against the Foresight obesity system map domains and nodes and develop a framework to report big data for obesity research. Opportunities and challenges associated with this new data approach to whole systems obesity research are discussed.MethodsExpert opinion from the ESRC Strategic Network for Obesity was harnessed in order to develop a data source reporting framework for obesity research. The framework was then tested on a range of data sources. In order to assess availability of data sources relevant to obesity research, a data mapping exercise against the Foresight obesity systems map domains and nodes was carried out.ResultsA reporting framework was developed to recommend the reporting of key information in line with these headings: Background; Elements; Exemplars; Content; Ownership; Aggregation; Sharing; Temporality (BEE-COAST). The new BEE-COAST framework was successfully applied to eight exemplar data sources from the UK. 80% coverage of the Foresight obesity systems map is possible using a wide range of big data sources. The remaining 20% were primarily biological measurements often captured by more traditional laboratory based research.ConclusionsBig data offer great potential across many domains of obesity research and need to be leveraged in conjunction with traditional data for societal benefit and health promotion.
| S-EPMC6291418 | biostudies-literature
Project description:Health advances are contingent on continuous development of new methods and approaches to foster data-driven discovery in the biomedical and clinical sciences. Open-science and team-based scientific discovery offer hope for tackling some of the difficult challenges associated with managing, modeling, and interpreting of large, complex, and multisource data. Translating raw observations into useful information and actionable knowledge depends on effective domain-independent reproducibility, area-specific replicability, data curation, analysis protocols, organization, management and sharing of health-related digital objects. This study expands the functionality and utility of an ensemble semi-supervised machine learning technique called Compressive Big Data Analytics (CBDA). Applied to high-dimensional data, CBDA (1) identifies salient features and key biomarkers enabling reliable and reproducible forecasting of binary, multinomial and continuous outcomes (i.e., feature mining); and (2) suggests the most accurate algorithms/models for predictive analytics of the observed data (i.e., model mining). The method relies on iterative subsampling, combines function optimization and statistical inference, and generates ensemble predictions for observed univariate outcomes. The novelty of this study is highlighted by a new and expanded set of CBDA features including (1) efficiently handling extremely large datasets (>100,000 cases and >1,000 features); (2) generalizing the internal and external validation steps; (3) expanding the set of base-learners for joint ensemble prediction; (4) introducing an automated selection of CBDA specifications; and (5) providing mechanisms to assess CBDA convergence, evaluate the prediction accuracy, and measure result consistency. To ground the mathematical model and the corresponding computational algorithm, CBDA 2.0 validation utilizes synthetic datasets as well as a population-wide census-like study. Specifically, an empirical validation of the CBDA technique is based on a translational health research using a large-scale clinical study (UK Biobank), which includes imaging, cognitive, and clinical assessment data. The UK Biobank archive presents several difficult challenges related to the aggregation, harmonization, modeling, and interrogation of the information. These problems are related to the complex longitudinal structure, variable heterogeneity, feature multicollinearity, incongruency, and missingness, as well as violations of classical parametric assumptions. Our results show the scalability, efficiency, and usability of CBDA to interrogate complex data into structural information leading to derived knowledge and translational action. Applying CBDA 2.0 to the UK Biobank case-study allows predicting various outcomes of interest, e.g., mood disorders and irritability, and suggests new and exciting avenues of evidence-based research in the context of identifying, tracking, and treating mental health and aging-related diseases. Following open-science principles, we share the entire end-to-end protocol, source-code, and results. This facilitates independent validation, result reproducibility, and team-based collaborative discovery.
| S-EPMC7455041 | biostudies-literature
Project description:A matched molecular pair (MMP) analysis was used to examine the change in melting point (MP) between pairs of similar molecules in a set of ∼275k compounds. We found many cases in which the change in MP (ΔMP) of compounds correlates with changes in functional groups. In line with the results of a previous study, correlations between ΔMP and simple molecular descriptors, such as the number of hydrogen bond donors, were identified. In using a larger dataset, covering a wider chemical space and range of melting points, we observed that this method remains stable and scales well with larger datasets. This MMP-based method could find use as a simple privacy-preserving technique to analyze large proprietary databases and share findings between participating research groups.
| S-EPMC5900986 | biostudies-literature
Project description:BackgroundLigand-based in silico target fishing can be used to identify the potential interacting target of bioactive ligands, which is useful for understanding the polypharmacology and safety profile of existing drugs. The underlying principle of the approach is that known bioactive ligands can be used as reference to predict the targets for a new compound.ResultsWe tested a pipeline enabling large-scale target fishing and drug repositioning, based on simple fingerprint similarity rankings with data fusion. A large library containing 533 drug relevant targets with 179,807 active ligands was compiled, where each target was defined by its ligand set. For a given query molecule, its target profile is generated by similarity searching against the ligand sets assigned to each target, for which individual searches utilizing multiple reference structures are then fused into a single ranking list representing the potential target interaction profile of the query compound. The proposed approach was validated by 10-fold cross validation and two external tests using data from DrugBank and Therapeutic Target Database (TTD). The use of the approach was further demonstrated with some examples concerning the drug repositioning and drug side-effects prediction. The promising results suggest that the proposed method is useful for not only finding promiscuous drugs for their new usages, but also predicting some important toxic liabilities.ConclusionsWith the rapid increasing volume and diversity of data concerning drug related targets and their ligands, the simple ligand-based target fishing approach would play an important role in assisting future drug design and discovery.
| S-EPMC4068908 | biostudies-literature
Project description:Secondary analyses of survey data collected from large probability samples of persons or establishments further scientific progress in many fields. The complex design features of these samples improve data collection efficiency, but also require analysts to account for these features when conducting analysis. Unfortunately, many secondary analysts from fields outside of statistics, biostatistics, and survey methodology do not have adequate training in this area, and as a result may apply incorrect statistical methods when analyzing these survey data sets. This in turn could lead to the publication of incorrect inferences based on the survey data that effectively negate the resources dedicated to these surveys. In this article, we build on the results of a preliminary meta-analysis of 100 peer-reviewed journal articles presenting analyses of data from a variety of national health surveys, which suggested that analytic errors may be extremely prevalent in these types of investigations. We first perform a meta-analysis of a stratified random sample of 145 additional research products analyzing survey data from the Scientists and Engineers Statistical Data System (SESTAT), which describes features of the U.S. Science and Engineering workforce, and examine trends in the prevalence of analytic error across the decades used to stratify the sample. We once again find that analytic errors appear to be quite prevalent in these studies. Next, we present several example analyses of real SESTAT data, and demonstrate that a failure to perform these analyses correctly can result in substantially biased estimates with standard errors that do not adequately reflect complex sample design features. Collectively, the results of this investigation suggest that reviewers of this type of research need to pay much closer attention to the analytic methods employed by researchers attempting to publish or present secondary analyses of survey data.
| S-EPMC4927119 | biostudies-literature
Project description:Today, people generate and store more data than ever before as they interact with both real and virtual environments. These digital traces of behavior and cognition offer cognitive scientists and psychologists an unprecedented opportunity to test theories outside the laboratory. Despite general excitement about big data and naturally occurring datasets among researchers, three "gaps" stand in the way of their wider adoption in theory-driven research: the imagination gap, the skills gap, and the culture gap. We outline an approach to bridging these three gaps while respecting our responsibilities to the public as participants in and consumers of the resulting research. To that end, we introduce Data on the Mind ( http://www.dataonthemind.org ), a community-focused initiative aimed at meeting the unprecedented challenges and opportunities of theory-driven research with big data and naturally occurring datasets. We argue that big data and naturally occurring datasets are most powerfully used to supplement-not supplant-traditional experimental paradigms in order to understand human behavior and cognition, and we highlight emerging ethical issues related to the collection, sharing, and use of these powerful datasets.
| S-EPMC5628193 | biostudies-other