scholarly journals mixOmics: an R package for ‘omics feature selection and multiple data integration

2017 ◽  
Author(s):  
Florian Rohart ◽  
Benoît Gautier ◽  
Amrit Singh ◽  
Kim-Anh Lê Cao
Keyword(s):  
Data Integration ◽  
Large Scale ◽  
Relevant Information ◽  
R Package ◽  
Biological Data ◽  
Molecular Signature ◽  
Single Type ◽  
Data Sets ◽  
Omics Data ◽  
Wide Range

AbstractThe advent of high throughput technologies has led to a wealth of publicly available ‘omics data coming from different sources, such as transcriptomics, proteomics, metabolomics. Combining such large-scale biological data sets can lead to the discovery of important biological insights, provided that relevant information can be extracted in a holistic manner. Current statistical approaches have been focusing on identifying small subsets of molecules (a ‘molecular signature’) to explain or predict biological conditions, but mainly for a single type of ‘omics. In addition, commonly used methods are univariate and consider each biological feature independently.We introducemixOmics, an R package dedicated to the multivariate analysis of biological data sets with a specific focus on data exploration, dimension reduction and visualisation. By adopting a system biology approach, the toolkit provides a wide range of methods that statistically integrate several data sets at once to probe relationships between heterogeneous ‘omics data sets. Our recent methods extend Projection to Latent Structure (PLS) models for discriminant analysis, for data integration across multiple ‘omics data or across independent studies, and for the identification of molecular signatures. We illustrate our latestmixOmicsintegrative frameworks for the multivariate analyses of ‘omics data available from the package.

scholarly journals Denoising large-scale biological data using network filters

2021 ◽  
Author(s):  
Andrew J Kavran ◽  
Aaron Clauset
Keyword(s):  
Large Scale ◽  
Synthetic Data ◽  
Interaction Network ◽  
Learning Task ◽  
Biological Data ◽  
Data Sets ◽  
Proteomics Data ◽  
Life History Variation ◽  
Wide Range ◽  
Underlying Processes

Abstract Background: Large-scale biological data sets are often contaminated by noise, which can impede accurate inferences about underlying processes. Such measurement noise can arise from endogenous biological factors like cell cycle and life history variation, and from exogenous technical factors like sample preparation and instrument variation.Results: We describe a general method for automatically reducing noise in large-scale biological data sets. This method uses an interaction network to identify groups of correlated or anti-correlated measurements that can be combined or “filtered” to better recover an underlying biological signal. Similar to the process of denoising an image, a single network filter may be applied to an entire system, or the system may be first decomposed into distinct modules and a different filter applied to each. Applied to synthetic data with known network structure and signal, network filters accurately reduce noise across a wide range of noise levels and structures. Applied to a machine learning task of predicting changes in human protein expression in healthy and cancerous tissues, network filtering prior to training increases accuracy up to 43% compared to using unfiltered data.Conclusions: Network filters are a general way to denoise biological data and can account for both correlation and anti-correlation between different measurements. Furthermore, we find that partitioning a network prior to filtering can significantly reduce errors in networks with heterogenous data and correlation patterns, and this approach outperforms existing diffusion based methods. Our results on proteomics data indicate the broad potential utility of network filters to applications in systems biology.

scholarly journals Denoising Large-Scale Biological Data Using Network Filters

2020 ◽  
Author(s):  
Andrew J Kavran ◽  
Aaron Clauset
Keyword(s):  
Large Scale ◽  
Synthetic Data ◽  
Interaction Network ◽  
Learning Task ◽  
Biological Data ◽  
Data Sets ◽  
Proteomics Data ◽  
Life History Variation ◽  
Wide Range ◽  
Underlying Processes

Abstract Background: Large-scale biological data sets, e.g., transcriptomic, proteomic, or ecological, are often contaminated by noise, which can impede accurate inferences about underlying processes. Such measurement noise can arise from endogenous biological factors like cell cycle and life history variation, and from exogenous technical factors like sample preparation and instrument variation. Results: We describe a general method for automatically reducing noise in large-scale biological data sets. This method uses an interaction network to identify groups of correlated or anti-correlated measurements that can be combined or “filtered” to better recover an underlying biological signal. Similar to the process of denoising an image, a single network filter may be applied to an entire system, or the system may be first decomposed into distinct modules and a different filter applied to each. Applied to synthetic data with known network structure and signal, network filters accurately reduce noise across a wide range of noise levels and structures. Applied to a machine learning task of predicting changes in human protein expression in healthy and cancerous tissues, network filtering prior to training increases accuracy up to 58% compared to using unfiltered data. Conclusions: Network filters are a general way to denoise biological data and can account for both correlation and anti-correlation between different measurements. Furthermore, we find that partitioning a network prior to filtering can significantly reduce errors in networks with heterogenous data and correlation patterns, andthis approach outperforms existing diffusion based methods. Our results on proteomics data indicate the broad potential utility of network filters to applications in systems biology.

scholarly journals Denoising large-scale biological data using network filters

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Andrew J. Kavran ◽  
Aaron Clauset
Keyword(s):  
Large Scale ◽  
Synthetic Data ◽  
Interaction Network ◽  
Learning Task ◽  
Biological Data ◽  
Data Sets ◽  
Proteomics Data ◽  
Life History Variation ◽  
Wide Range ◽  
Underlying Processes

Abstract Background Large-scale biological data sets are often contaminated by noise, which can impede accurate inferences about underlying processes. Such measurement noise can arise from endogenous biological factors like cell cycle and life history variation, and from exogenous technical factors like sample preparation and instrument variation. Results We describe a general method for automatically reducing noise in large-scale biological data sets. This method uses an interaction network to identify groups of correlated or anti-correlated measurements that can be combined or “filtered” to better recover an underlying biological signal. Similar to the process of denoising an image, a single network filter may be applied to an entire system, or the system may be first decomposed into distinct modules and a different filter applied to each. Applied to synthetic data with known network structure and signal, network filters accurately reduce noise across a wide range of noise levels and structures. Applied to a machine learning task of predicting changes in human protein expression in healthy and cancerous tissues, network filtering prior to training increases accuracy up to 43% compared to using unfiltered data. Conclusions Network filters are a general way to denoise biological data and can account for both correlation and anti-correlation between different measurements. Furthermore, we find that partitioning a network prior to filtering can significantly reduce errors in networks with heterogenous data and correlation patterns, and this approach outperforms existing diffusion based methods. Our results on proteomics data indicate the broad potential utility of network filters to applications in systems biology.

scholarly journals Denoising large-scale biological data using network filters

2020 ◽  
Author(s):  
Andrew J. Kavran ◽  
Aaron Clauset
Keyword(s):  
Large Scale ◽  
Interaction Network ◽  
Biological Data ◽  
Neuron Activity ◽  
Data Sets ◽  
Correlated Noise ◽  
Life History Variation ◽  
Real World Data ◽  
Brain Connectome ◽  
Wide Range

Large-scale biological data sets, e.g., transcriptomic, proteomic, or ecological, are often contaminated by noise, which can impede accurate inferences about underlying processes. Such measurement noise can arise from endogenous biological factors like cell cycle and life history variation, and from exogenous technical factors like sample preparation and instrument variation. Here we describe a general method for automatically reducing noise in large-scale biological data sets. This method uses an interaction network to identify groups of correlated or anti-correlated measurements that can be combined or “filtered” to better recover an underlying biological signal. Similar to the process of denoising an image, a single network filter may be applied to an entire system, or the system may be first decomposed into distinct modules and a different filter applied to each. Applied to synthetic data with known network structure and signal, network filters accurately reduce noise across a wide range of noise levels and structures. Applied to a machine learning task of predicting changes in human protein expression in healthy and cancerous tissues, network filtering prior to training increases accuracy up to 58% compared to using unfiltered data. These results indicate the broad potential utility of network-based filters to applications in systems biology.Author SummarySystem-wide measurements of many biological signals, whether derived from molecules, cells, or entire organisms, are often noisy. Removing or mitigating this noise prior to analysis can improve our understanding and predictions of biological phenomena. We describe a general way to denoise biological data that can account for both correlation and anti-correlation between different measurements. These “network filters” take as input a set of biological measurements, e.g., metabolite concentration, animal traits, neuron activity, or gene expression, and a network of how those measurements are biologically related, e.g., a metabolic network, food web, brain connectome, or protein-protein interaction network. Measurements are then “filtered” for correlated or anti-correlated noise using a set of other measurements that are identified using the network. We investigate the accuracy of these filters in synthetic and real-world data sets, and find that they can substantially reduce noise of different levels and structure. By denoising large-scale biological data sets, network filters have the potential to improve the analysis of many types of biological data.

scholarly journals MUREN: a robust and multi-reference approach of RNA-seq transcript normalization

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Yance Feng ◽  
Lei M. Li
Keyword(s):  
Biological Significance ◽  
Housekeeping Genes ◽  
R Package ◽  
Data Sets ◽  
Statistical Regression ◽  
Rna Seq ◽  
Least Trimmed Squares ◽  
Standard Data ◽  
Wide Range ◽  
Multiple References

Abstract Background Normalization of RNA-seq data aims at identifying biological expression differentiation between samples by removing the effects of unwanted confounding factors. Explicitly or implicitly, the justification of normalization requires a set of housekeeping genes. However, the existence of housekeeping genes common for a very large collection of samples, especially under a wide range of conditions, is questionable. Results We propose to carry out pairwise normalization with respect to multiple references, selected from representative samples. Then the pairwise intermediates are integrated based on a linear model that adjusts the reference effects. Motivated by the notion of housekeeping genes and their statistical counterparts, we adopt the robust least trimmed squares regression in pairwise normalization. The proposed method (MUREN) is compared with other existing tools on some standard data sets. The goodness of normalization emphasizes on preserving possible asymmetric differentiation, whose biological significance is exemplified by a single cell data of cell cycle. MUREN is implemented as an R package. The code under license GPL-3 is available on the github platform: github.com/hippo-yf/MUREN and on the conda platform: anaconda.org/hippo-yf/r-muren. Conclusions MUREN performs the RNA-seq normalization using a two-step statistical regression induced from a general principle. We propose that the densities of pairwise differentiations are used to evaluate the goodness of normalization. MUREN adjusts the mode of differentiation toward zero while preserving the skewness due to biological asymmetric differentiation. Moreover, by robustly integrating pre-normalized counts with respect to multiple references, MUREN is immune to individual outlier samples.

scholarly journals Comparative Cladistics: Fossils, Morphological Data Partitions and Lost Branches in the Fossil Tree of Life

2017 ◽  
Author(s):  
Ross Mounce
Keyword(s):  
Large Scale ◽  
Phylogenetic Signal ◽  
Scientific Progress ◽  
Molecular Data ◽  
Morphological Data ◽  
Data Sets ◽  
Phylogenetic Groups ◽  
Use Of Data ◽  
Wide Range ◽  
Cladistic Analyses

In this thesis I attempt to gather together a wide range of cladistic analyses of fossil and extant taxa representing a diverse array of phylogenetic groups. I use this data to quantitatively compare the effect of fossil taxa relative to extant taxa in terms of support for relationships, number of most parsimonious trees (MPTs) and leaf stability. In line with previous studies I find that the effects of fossil taxa are seldom different to extant taxa – although I highlight some interesting exceptions. I also use this data to compare the phylogenetic signal within vertebrate morphological data sets, by choosing to compare cranial data to postcranial data. Comparisons between molecular data and morphological data have been previously well explored, as have signals between different molecular loci. But comparative signal within morphological data sets is much less commonly characterized and certainly not across a wide array of clades. With this analysis I show that there are many studies in which the evidence provided by cranial data appears to be be significantly incongruent with the postcranial data – more than one would expect to see just by the effect of chance and noise alone. I devise and implement a modification to a rarely used measure of homoplasy that will hopefully encourage its wider usage. Previously it had some undesirable bias associated with the distribution of missing data in a dataset, but my modification controls for this. I also take an in-depth and extensive review of the ILD test, noting it is often misused or reported poorly, even in recent studies. Finally, in attempting to collect data and metadata on a large scale, I uncovered inefficiencies in the research publication system that obstruct re-use of data and scientific progress. I highlight the importance of replication and reproducibility – even simple reanalysis of high profile papers can turn up some very different results. Data is highly valuable and thus it must be retained and made available for further re-use to maximize the overall return on research investment.

scholarly journals Usage and Scaling of an Open-Source Spiking Multi-Area Model of Monkey Cortex

2021 ◽  
pp. 47-59
Author(s):  
Sacha J. van Albada ◽  
Jari Pronold ◽  
Alexander van Meegen ◽  
Markus Diesmann
Keyword(s):  
Open Source ◽  
Large Scale ◽  
Network Models ◽  
Macaque Monkey ◽  
Source Model ◽  
Model Specification ◽  
Data Sets ◽  
Neural Network Models ◽  
Wide Range ◽  
Ict Infrastructure

AbstractWe are entering an age of ‘big’ computational neuroscience, in which neural network models are increasing in size and in numbers of underlying data sets. Consolidating the zoo of models into large-scale models simultaneously consistent with a wide range of data is only possible through the effort of large teams, which can be spread across multiple research institutions. To ensure that computational neuroscientists can build on each other’s work, it is important to make models publicly available as well-documented code. This chapter describes such an open-source model, which relates the connectivity structure of all vision-related cortical areas of the macaque monkey with their resting-state dynamics. We give a brief overview of how to use the executable model specification, which employs NEST as simulation engine, and show its runtime scaling. The solutions found serve as an example for organizing the workflow of future models from the raw experimental data to the visualization of the results, expose the challenges, and give guidance for the construction of an ICT infrastructure for neuroscience.

Characterising RDF data sets

2017 ◽  
Vol 44 (2) ◽  
pp. 203-229 ◽  
Author(s):  
Javier D Fernández ◽  
Miguel A Martínez-Prieto ◽  
Pablo de la Fuente Redondo ◽  
Claudio Gutiérrez
Keyword(s):  
Data Structures ◽  
Large Scale ◽  
Open Data ◽  
Structural Features ◽  
Data Sets ◽  
Data Set ◽  
Wide Range ◽  
Rdf Data ◽  
Description Framework ◽  
Resource Description

The publication of semantic web data, commonly represented in Resource Description Framework (RDF), has experienced outstanding growth over the last few years. Data from all fields of knowledge are shared publicly and interconnected in active initiatives such as Linked Open Data. However, despite the increasing availability of applications managing large-scale RDF information such as RDF stores and reasoning tools, little attention has been given to the structural features emerging in real-world RDF data. Our work addresses this issue by proposing specific metrics to characterise RDF data. We specifically focus on revealing the redundancy of each data set, as well as common structural patterns. We evaluate the proposed metrics on several data sets, which cover a wide range of designs and models. Our findings provide a basis for more efficient RDF data structures, indexes and compressors.

Analysis and Integration of Biological Data

2010 ◽  
pp. 287-314
Author(s):  
Diego Milone ◽  
Georgina Stegmayer ◽  
Matías Gerard ◽  
Laura Kamenetzky ◽  
Mariana López ◽  
...  
Keyword(s):  
Data Integration ◽  
New Technologies ◽  
Biological Significance ◽  
Neural Model ◽  
Biological Data ◽  
Data Sets ◽  
Clustering Methods ◽  
Gene Expressions ◽  
Internal Coherence ◽  
Biological Data Integration

The volume of information derived from post genomic technologies is rapidly increasing. Due to the amount of involved data, novel computational methods are needed for the analysis and knowledge discovery into the massive data sets produced by these new technologies. Furthermore, data integration is also gaining attention for merging signals from different sources in order to discover unknown relations. This chapter presents a pipeline for biological data integration and discovery of a priori unknown relationships between gene expressions and metabolite accumulations. In this pipeline, two standard clustering methods are compared against a novel neural network approach. The neural model provides a simple visualization interface for identification of coordinated patterns variations, independently of the number of produced clusters. Several quality measurements have been defined for the evaluation of the clustering results obtained on a case study involving transcriptomic and metabolomic profiles from tomato fruits. Moreover, a method is proposed for the evaluation of the biological significance of the clusters found. The neural model has shown a high performance in most of the quality measures, with internal coherence in all the identified clusters and better visualization capabilities.

scholarly journals Challenges in the Integration of Omics and Non-Omics Data

Genes ◽  
2019 ◽  
Vol 10 (3) ◽  
pp. 238 ◽  
Author(s):  
Evangelina López de Maturana ◽  
Lola Alonso ◽  
Pablo Alarcón ◽  
Isabel Adoración Martín-Antoniano ◽  
Silvia Pineda ◽  
...  
Keyword(s):  
Data Integration ◽  
Large Scale ◽  
Predictive Ability ◽  
Epidemiological Data ◽  
Joint Modeling ◽  
Epidemiological Studies ◽  
Omics Data ◽  
Data Set ◽  
Health Domains ◽  
Analytical Strategies

Omics data integration is already a reality. However, few omics-based algorithms show enough predictive ability to be implemented into clinics or public health domains. Clinical/epidemiological data tend to explain most of the variation of health-related traits, and its joint modeling with omics data is crucial to increase the algorithm’s predictive ability. Only a small number of published studies performed a “real” integration of omics and non-omics (OnO) data, mainly to predict cancer outcomes. Challenges in OnO data integration regard the nature and heterogeneity of non-omics data, the possibility of integrating large-scale non-omics data with high-throughput omics data, the relationship between OnO data (i.e., ascertainment bias), the presence of interactions, the fairness of the models, and the presence of subphenotypes. These challenges demand the development and application of new analysis strategies to integrate OnO data. In this contribution we discuss different attempts of OnO data integration in clinical and epidemiological studies. Most of the reviewed papers considered only one type of omics data set, mainly RNA expression data. All selected papers incorporated non-omics data in a low-dimensionality fashion. The integrative strategies used in the identified papers adopted three modeling methods: Independent, conditional, and joint modeling. This review presents, discusses, and proposes integrative analytical strategies towards OnO data integration.

Sign in / Sign up

Export Citation Format

Share Document