Detecting representative data and generating synthetic samples to improve learning accuracy with imbalanced data sets.
Ontology highlight
ABSTRACT:
SUBMITTER: Li DC
PROVIDER: S-EPMC5542532 | biostudies-other | 2017
REPOSITORIES: biostudies-other
Similar Datasets
Project description:High-Throughput technologies provide genomic and trascriptomic data that are suitable for biomarker detection for classification purposes. However, the high dimension of the output of such technologies and the characteristics of the data sets analysed represent an issue for the classification task. Here we present a new feature selection method based on three steps to detect class-specific biomarkers in case of high-dimensional data sets. The first step detects the differentially expressed genes according to the experimental conditions tested in the experimental design, the second step filters out the features with low discriminative power and the third step detects the class-specific features and defines the final biomarker as the union of the class-specific features. The proposed procedure is tested on two microarray datasets, one characterized by a strong imbalance between the size of classes and the other one where the size of classes is perfectly balanced. We show that, using the proposed feature selection procedure, the classification performances of a Support Vector Machine on the imbalanced data set reach a 82% whereas other methods do not exceed 73%. Furthermore, in case of perfectly balanced dataset, the classification performances are comparable with other methods. Finally, the Gene Ontology enrichments performed on the signatures selected with the proposed pipeline, confirm the biological relevance of our methodology. The download of the package with the implementation of Peculiar Genes Selection, 'PGS', is available for R users at: http://github.com/mbeccuti/PGS.
| S-EPMC5555681 | biostudies-literature
Project description:The Protein Data Bank currently contains about 600 data sets of three-dimensional protein coordinates determined by X-ray crystallography or NMR. There is considerable redundancy in the data base, as many protein pairs are identical or very similar in sequence. However, statistical analyses of protein sequence-structure relations require nonredundant data. We have developed two algorithms to extract from the data base representative sets of protein chains with maximum coverage and minimum redundancy. The first algorithm focuses on optimizing a particular property of the selected proteins and works by successive selection of proteins from an ordered list and exclusion of all neighbors of each selected protein. The other algorithm aims at maximizing the size of the selected set and works by successive thinning out of clusters of similar proteins. Both algorithms are generally applicable to other data bases in which criteria of similarity can be defined and relate to problems in graph theory. The largest nonredundant set extracted from the current release of the Protein Data Bank has 155 protein chains. In this set, no two proteins have sequence similarity higher than a certain cutoff (30% identical residues for aligned subsequences longer than 80 residues), yet all structurally unique protein families are represented. Periodically updated lists of representative data sets are available by electronic mail from the file server "netserv@embl-heidelberg.de." The selection may be useful in statistical approaches to protein folding as well as in the analysis and documentation of the known spectrum of three-dimensional protein structures.
| S-EPMC2142204 | biostudies-other
Project description:Identifying interesting relationships between pairs of variables in large data sets is increasingly important. Here, we present a measure of dependence for two-variable relationships: the maximal information coefficient (MIC). MIC captures a wide range of associations both functional and not, and for functional relationships provides a score that roughly equals the coefficient of determination (R(2)) of the data relative to the regression function. MIC belongs to a larger class of maximal information-based nonparametric exploration (MINE) statistics for identifying and classifying relationships. We apply MIC and MINE to data sets in global health, gene expression, major-league baseball, and the human gut microbiota and identify known and novel relationships.
| S-EPMC3325791 | biostudies-other
Project description:MotivationDespite numerous RNA-seq samples available at large databases, most RNA-seq analysis tools are evaluated on a limited number of RNA-seq samples. This drives a need for methods to select a representative subset from all available RNA-seq samples to facilitate comprehensive, unbiased evaluation of bioinformatics tools. In sequence-based approaches for representative set selection (e.g. a k-mer counting approach that selects a subset based on k-mer similarities between RNA-seq samples), because of the large numbers of available RNA-seq samples and of k-mers/sequences in each sample, computing the full similarity matrix using k-mers/sequences for the entire set of RNA-seq samples in a large database (e.g. the SRA) has memory and runtime challenges; this makes direct representative set selection infeasible with limited computing resources.ResultsWe developed a novel computational method called 'hierarchical representative set selection' to handle this challenge. Hierarchical representative set selection is a divide-and-conquer-like algorithm that breaks representative set selection into sub-selections and hierarchically selects representative samples through multiple levels. We demonstrate that hierarchical representative set selection can achieve summarization quality close to that of direct representative set selection, while largely reducing runtime and memory requirements of computing the full similarity matrix (up to 8.4× runtime reduction and 5.35× memory reduction for 10 000 and 12 000 samples respectively that could be practically run with direct subset selection). We show that hierarchical representative set selection substantially outperforms random sampling on the entire SRA set of RNA-seq samples, making it a practical solution to representative set selection on large databases like the SRA.Availability and implementationThe code is available at https://github.com/Kingsford-Group/hierrepsetselection and https://github.com/Kingsford-Group/jellyfishsim.Supplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC8275344 | biostudies-literature
Project description:Getting access to administrative health data for research purposes is a difficult and time-consuming process due to increasingly demanding privacy regulations. An alternative method for sharing administrative health data would be to share synthetic datasets where the records do not correspond to real individuals, but the patterns and relationships seen in the data are reproduced. This paper assesses the feasibility of generating synthetic administrative health data using a recurrent deep learning model. Our data comes from 120,000 individuals from Alberta Health's administrative health database. We assess how similar our synthetic data is to the real data using utility assessments that assess the structure and general patterns in the data as well as by recreating a specific analysis in the real data commonly applied to this type of administrative health data. We also assess the privacy risks associated with the use of this synthetic dataset. Generic utility assessments that used Hellinger distance to quantify the difference in distributions between real and synthetic datasets for event types (0.027), attributes (mean 0.0417), Markov transition matrices (order 1 mean absolute difference: 0.0896, sd: 0.159; order 2: mean Hellinger distance 0.2195, sd: 0.2724), the Hellinger distance between the joint distributions was 0.352, and the similarity of random cohorts generated from real and synthetic data had a mean Hellinger distance of 0.3 and mean Euclidean distance of 0.064, indicating small differences between the distributions in the real data and the synthetic data. By applying a realistic analysis to both real and synthetic datasets, Cox regression hazard ratios achieved a mean confidence interval overlap of 68% for adjusted hazard ratios among 5 key outcomes of interest, indicating synthetic data produces similar analytic results to real data. The privacy assessment concluded that the attribution disclosure risk associated with this synthetic dataset was substantially less than the typical 0.09 acceptable risk threshold. Based on these metrics our results show that our synthetic data is suitably similar to the real data and could be shared for research purposes thereby alleviating concerns associated with the sharing of real data in some circumstances.
| S-EPMC10034254 | biostudies-literature
Project description:Chromatin immunoprecipitation DNA-sequencing (ChIP-seq) with and without designed internal controls.
2023-05-23 | GSE230475 | GEO
Project description:In high-throughput molecular profiling studies, genotype labels can be wrongly assigned at various experimental steps; the resulting mislabeled samples seriously reduce the power to detect the genetic basis of phenotypic variation. We have developed an approach to detect potential mislabeling, recover the "ideal" genotype and identify "best-matched" labels for mislabeled samples. On average, we identified 4% of samples as mislabeled in eight published datasets, highlighting the necessity of applying a "data cleaning" step before standard data analysis.
| S-EPMC5305221 | biostudies-literature
Project description:IntroductionFederated Learning (FL) is a decentralised approach to training statistical models, where training is performed across multiple clients, producing one global model. Since the training data remains with each local client and is not shared or exchanged with other clients the use of FL may reduce privacy and security risks (compared to methods where multiple data sources are pooled) and can also address data access and heterogeneity problems. Synthetic data is artificially generated data that has the same structure and statistical properties as the original but that does not contain any of the original data records, therefore minimising disclosure risk. Using FL to produce synthetic data (which we refer to as "federated synthesis") has the potential to combine data from multiple clients without compromising privacy, allowing access to data that may otherwise be inaccessible in its raw format.ObjectivesThe objective was to review current research and practices for using FL to generate synthetic data and determine the extent to which research has been undertaken, the methods and evaluation practices used, and any research gaps.MethodsA scoping review was conducted to systematically map and describe the published literature on the use of FL to generate synthetic data. Relevant studies were identified through online databases and the findings are described, grouped, and summarised. Information extracted included article characteristics, documenting the type of data that is synthesised, the model architecture and the methods (if any) used to evaluate utility and privacy risk.ResultsA total of 69 articles were included in the scoping review; all were published between 2018 and 2023 with two thirds (46) in 2022. 30% (21) were focussed on synthetic data generation as the main model output (with 6 of these generating tabular data), whereas 59% (41) focussed on data augmentation. Of the 21 performing federated synthesis, all used deep learning methods (predominantly Generative Adversarial Networks) to generate the synthetic data.ConclusionsFederated synthesis is in its early days but shows promise as a method that can construct a global synthetic dataset without sharing any of the local client data. As a field in its infancy there are areas to explore in terms of the privacy risk associated with the various methods proposed, and more generally in how we measure those risks.
| S-EPMC10898505 | biostudies-literature
Project description:BackgroundWith the completion of the HapMap project, a variety of computational algorithms and tools have been proposed for haplotype inference, tag SNP selection and genome-wide association studies. Simulated data are commonly used in evaluating these new developed approaches. In addition to simulations based on population models, empirical data generated by perturbing real data, has also been used because it may inherit specific properties from real data. However, there is no tool that is publicly available to generate large scale simulated variation data by taking into account knowledge from the HapMap project.ResultsA computer program (gs) was developed to quickly generate a large number of samples based on real data that are useful for a variety of purposes, including evaluating methods for haplotype inference, tag SNP selection and association studies. Two approaches have been implemented to generate dense SNP haplotype/genotype data that share similar local linkage disequilibrium (LD) patterns as those in human populations. The first approach takes haplotype pairs from samples as inputs, and the second approach takes patterns of haplotype block structures as inputs. Both quantitative and qualitative traits have been incorporated in the program. Phenotypes are generated based on a disease model, or based on the effect of a quantitative trait nucleotide, both of which can be specified by users. In addition to single-locus disease models, two-locus disease models have also been implemented that can incorporate any degree of epistasis. Users are allowed to specify all nine parameters in a 3 x 3 penetrance table. For several commonly used two-locus disease models, the program can automatically calculate penetrances based on the population prevalence and marginal effects of a disease that users can conveniently specify.ConclusionThe program gs can effectively generate large scale genetic and phenotypic variation data that can be used for evaluating new developed approaches. It is freely available from the authors' web site at http://www.eecs.case.edu/~jxl175/gs.html.
| S-EPMC2375120 | biostudies-literature
Project description:Selecting a non-redundant representative subset of sequences is a common step in many bioinformatics workflows, such as the creation of non-redundant training sets for sequence and structural models or selection of "operational taxonomic units" from metagenomics data. Previous methods for this task, such as CD-HIT, PISCES, and UCLUST, apply a heuristic threshold-based algorithm that has no theoretical guarantees. We propose a new approach based on submodular optimization. Submodular optimization, a discrete analogue to continuous convex optimization, has been used with great success for other representative set selection problems. We demonstrate that the submodular optimization approach results in representative protein sequence subsets with greater structural diversity than sets chosen by existing methods, using as a gold standard the SCOPe library of protein domain structures. In this setting, submodular optimization consistently yields protein sequence subsets that include more SCOPe domain families than sets of the same size selected by competing approaches. We also show how the optimization framework allows us to design a mixture objective function that performs well for both large and small representative sets. The framework we describe is the best possible in polynomial time (under some assumptions), and it is flexible and intuitive because it applies a suite of generic methods to optimize one of a variety of objective functions.
| S-EPMC5835207 | biostudies-literature