scholarly journals Identification of disease-associated loci using machine learning for genotype and network data integration

2019 ◽  
Vol 35 (24) ◽  
pp. 5182-5190 ◽  
Author(s):  
Luis G Leal ◽  
Alessia David ◽  
Marjo-Riita Jarvelin ◽  
Sylvain Sebert ◽  
Minna Männikkö ◽  
...  
Keyword(s):  
Machine Learning ◽  
Gene Networks ◽  
Association Studies ◽  
R Package ◽  
Biological Data ◽  
Machine Learning Algorithms ◽  
Supplementary Information ◽  
Genome Wide Association Studies ◽  
Omics Data ◽  
Missing Heritability

Abstract Motivation Integration of different omics data could markedly help to identify biological signatures, understand the missing heritability of complex diseases and ultimately achieve personalized medicine. Standard regression models used in Genome-Wide Association Studies (GWAS) identify loci with a strong effect size, whereas GWAS meta-analyses are often needed to capture weak loci contributing to the missing heritability. Development of novel machine learning algorithms for merging genotype data with other omics data is highly needed as it could enhance the prioritization of weak loci. Results We developed cNMTF (corrected non-negative matrix tri-factorization), an integrative algorithm based on clustering techniques of biological data. This method assesses the inter-relatedness between genotypes, phenotypes, the damaging effect of the variants and gene networks in order to identify loci-trait associations. cNMTF was used to prioritize genes associated with lipid traits in two population cohorts. We replicated 129 genes reported in GWAS world-wide and provided evidence that supports 85% of our findings (226 out of 265 genes), including recent associations in literature (NLGN1), regulators of lipid metabolism (DAB1) and pleiotropic genes for lipid traits (CARM1). Moreover, cNMTF performed efficiently against strong population structures by accounting for the individuals’ ancestry. As the method is flexible in the incorporation of diverse omics data sources, it can be easily adapted to the user’s research needs. Availability and implementation An R package (cnmtf) is available at https://lgl15.github.io/cnmtf_web/index.html. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals “Guilt by association” is not competitive with genetic association for identifying autism risk genes

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Margot Gunning ◽  
Paul Pavlidis
Keyword(s):  
Machine Learning ◽  
Genetic Association ◽  
Gene Networks ◽  
Rare Variants ◽  
Association Studies ◽  
Genetic Disorders ◽  
Autism Spectrum ◽  
Biological Data ◽  
Disease Genes ◽  
Risk Genes

AbstractDiscovering genes involved in complex human genetic disorders is a major challenge. Many have suggested that machine learning (ML) algorithms using gene networks can be used to supplement traditional genetic association-based approaches to predict or prioritize disease genes. However, questions have been raised about the utility of ML methods for this type of task due to biases within the data, and poor real-world performance. Using autism spectrum disorder (ASD) as a test case, we sought to investigate the question: can machine learning aid in the discovery of disease genes? We collected 13 published ASD gene prioritization studies and evaluated their performance using known and novel high-confidence ASD genes. We also investigated their biases towards generic gene annotations, like number of association publications. We found that ML methods which do not incorporate genetics information have limited utility for prioritization of ASD risk genes. These studies perform at a comparable level to generic measures of likelihood for the involvement of genes in any condition, and do not out-perform genetic association studies. Future efforts to discover disease genes should be focused on developing and validating statistical models for genetic association, specifically for association between rare variants and disease, rather than developing complex machine learning methods using complex heterogeneous biological data with unknown reliability.

scholarly journals bGWAS: an R package to perform Bayesian genome wide association studies

2020 ◽  
Vol 36 (15) ◽  
pp. 4374-4376
Author(s):  
Ninon Mounier ◽  
Zoltán Kutalik
Keyword(s):  
Mendelian Randomization ◽  
Causal Effect ◽  
Association Studies ◽  
R Package ◽  
Genome Wide Association ◽  
Supplementary Information ◽  
Genome Wide Association Studies ◽  
Biological Mechanisms ◽  
Genome Wide ◽  
Related Risk

Abstract Summary Increasing sample size is not the only strategy to improve discovery in Genome Wide Association Studies (GWASs) and we propose here an approach that leverages published studies of related traits to improve inference. Our Bayesian GWAS method derives informative prior effects by leveraging GWASs of related risk factors and their causal effect estimates on the focal trait using multivariable Mendelian randomization. These prior effects are combined with the observed effects to yield Bayes Factors, posterior and direct effects. The approach not only increases power, but also has the potential to dissect direct and indirect biological mechanisms. Availability and implementation bGWAS package is freely available under a GPL-2 License, and can be accessed, alongside with user guides and tutorials, from https://github.com/n-mounier/bGWAS. Supplementary information Supplementary data are available at Bioinformatics online.

Gene-based association tests using GWAS summary statistics

2019 ◽  
Vol 35 (19) ◽  
pp. 3701-3708 ◽  
Author(s):  
Gulnara R Svishcheva ◽  
Nadezhda M Belonogova ◽  
Irina V Zorkoltseva ◽  
Anatoly V Kirichenko ◽  
Tatiana I Axenovich
Keyword(s):  
Association Analysis ◽  
Association Studies ◽  
R Package ◽  
Supplementary Information ◽  
Genome Wide Association Studies ◽  
Summary Statistics ◽  
New Material ◽  
Artery Disease ◽  
The Many ◽  
Functional Linear Regression

Abstract Motivation A huge number of genome-wide association studies (GWAS) summary statistics freely available in databases provide a new material for gene-based association analysis aimed at identifying rare genetic variants. Only a few of the many popular gene-based methods developed for individual genotype and phenotype data are adapted for the practical use of the GWAS summary statistics as input. Results We analytically prove and numerically illustrate that all popular powerful methods developed for gene-based association analysis of individual phenotype and genotype data can be modified to utilize GWAS summary statistics. We have modified and implemented all of the popular methods, including burden and kernel machine-based tests, multiple and functional linear regression, principal components analysis and others, in the R package sumFREGAT. Using real summary statistics for coronary artery disease, we show that the new package is able to detect genes not found by the existing packages. Availability and implementation The R package sumFREGAT is freely and publicly available at: https://CRAN.R-project.org/package=sumFREGAT. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Multi-SNP mediation intersection-union test

2019 ◽  
Vol 35 (22) ◽  
pp. 4724-4729 ◽  
Author(s):  
Wujuan Zhong ◽  
Cassandra N Spracklen ◽  
Karen L Mohlke ◽  
Xiaojing Zheng ◽  
Jason Fine ◽  
...  
Keyword(s):  
Association Studies ◽  
R Package ◽  
Alternative Methods ◽  
Supplementary Information ◽  
Genome Wide Association Studies ◽  
Mediation Effects ◽  
Coding Regions ◽  
Genome Wide ◽  
Plasma Adiponectin Level ◽  
Intersection Union Test

Abstract Summary Tens of thousands of reproducibly identified GWAS (Genome-Wide Association Studies) variants, with the vast majority falling in non-coding regions resulting in no eventual protein products, call urgently for mechanistic interpretations. Although numerous methods exist, there are few, if any methods, for simultaneously testing the mediation effects of multiple correlated SNPs via some mediator (e.g. the expression of a gene in the neighborhood) on phenotypic outcome. We propose multi-SNP mediation intersection-union test (SMUT) to fill in this methodological gap. Our extensive simulations demonstrate the validity of SMUT as well as substantial, up to 92%, power gains over alternative methods. In addition, SMUT confirmed known mediators in a real dataset of Finns for plasma adiponectin level, which were missed by many alternative methods. We believe SMUT will become a useful tool to generate mechanistic hypotheses underlying GWAS variants, facilitating functional follow-up. Availability and implementation The R package SMUT is publicly available from CRAN at https://CRAN.R-project.org/package=SMUT. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Multiple similarly effective solutions exist for biomedical feature selection and classification problems

2017 ◽  
Vol 7 (1) ◽  
Author(s):  
Jiamei Liu ◽  
Cheng Xu ◽  
Weifeng Yang ◽  
Yayun Shu ◽  
Weiwei Zheng ◽  
...  
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Association Studies ◽  
Binary Classification ◽  
Learning Algorithms ◽  
Optimal Solution ◽  
Machine Learning Algorithms ◽  
Disease Classification ◽  
Genome Wide Association Studies ◽  
Classification Problems

Abstract Binary classification is a widely employed problem to facilitate the decisions on various biomedical big data questions, such as clinical drug trials between treated participants and controls, and genome-wide association studies (GWASs) between participants with or without a phenotype. A machine learning model is trained for this purpose by optimizing the power of discriminating samples from two groups. However, most of the classification algorithms tend to generate one locally optimal solution according to the input dataset and the mathematical presumptions of the dataset. Here we demonstrated from the aspects of both disease classification and feature selection that multiple different solutions may have similar classification performances. So the existing machine learning algorithms may have ignored a horde of fishes by catching only a good one. Since most of the existing machine learning algorithms generate a solution by optimizing a mathematical goal, it may be essential for understanding the biological mechanisms for the investigated classification question, by considering both the generated solution and the ignored ones.

scholarly journals MungeSumstats: A Bioconductor package for the standardisation and quality control of many GWAS summary statistics

2021 ◽  
Author(s):  
Alan E Murphy ◽  
Brian M Schilder ◽  
Nathan G Skene
Keyword(s):  
Quality Control ◽  
Association Studies ◽  
Meta Analysis ◽  
Genetic Research ◽  
Secondary Analysis ◽  
R Package ◽  
Supplementary Information ◽  
Genome Wide Association Studies ◽  
Summary Statistics ◽  
Summary Statistic

Abstract Motivation Genome-wide association studies (GWAS) summary statistics have popularised and accelerated genetic research. However, a lack of standardisation of the file formats used has proven problematic when running secondary analysis tools or performing meta-analysis studies. Results To address this issue, we have developed MungeSumstats, a Bioconductor R package for the standardisation and quality control of GWAS summary statistics. MungeSumstats can handle the most common summary statistic formats, including variant call format (VCF) producing a reformatted, standardised, tabular summary statistic file, VCF or R native data object. Availability MungeSumstats is available on Bioconductor (v 3.13) and can also be found on Github at: https://neurogenomics.github.io/MungeSumstats Supplementary information The analysis deriving the most common summary statistic formats is available at: https://al-murphy.github.io/SumstatFormats

scholarly journals VarGen: an R package for disease-associated variant discovery and annotation

2019 ◽  
Vol 36 (8) ◽  
pp. 2626-2627
Author(s):  
Corentin Molitor ◽  
Matt Brember ◽  
Fady Mohareb
Keyword(s):  
Association Studies ◽  
Genetic Disorders ◽  
R Package ◽  
Tissue Expression ◽  
Mendelian Inheritance ◽  
Supplementary Information ◽  
Genome Wide Association Studies ◽  
Variant Discovery ◽  
Genome Wide ◽  
High Quality Information

Abstract Summary Over the past decade, there has been an exponential increase in the amount of disease-related genomic data available in public databases. However, this high-quality information is spread across independent sources and researchers often need to access these separately. Hence, there is a growing need for tools that gather and compile this information in an easy and automated manner. Here, we present ‘VarGen’, an easy-to-use, customizable R package that fetches, annotates and rank variants related to diseases and genetic disorders, using a collection public databases (viz. Online Mendelian Inheritance in Man, the Functional Annotation of the Mammalian genome 5, the Genotype-Tissue Expression and the Genome Wide Association Studies catalog). This package is also capable of annotating these variants to identify the most impactful ones. We expect that this tool will benefit the research of variant-disease relationships. Availability and implementation VarGen is open-source and freely available via GitHub: https://github.com/MCorentin/VarGen. The software is implemented as an R package and is supported on Linux, MacOS and Windows. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Can machine learning aid in identifying disease genes? The case of autism spectrum disorder

2020 ◽  
Author(s):  
Margot Gunning ◽  
Paul Pavlidis
Keyword(s):  
Machine Learning ◽  
Autism Spectrum Disorder ◽  
Genetic Association ◽  
Gene Networks ◽  
Rare Variants ◽  
Association Studies ◽  
Autism Spectrum ◽  
Biological Data ◽  
Spectrum Disorder ◽  
Disease Genes

AbstractDiscovering genes involved in complex human genetic disorders is a major challenge. Many have suggested that machine learning (ML) algorithms using gene networks can be used to supplement traditional genetic association-based approaches to predict or prioritize disease genes. However, questions have been raised about the utility of ML methods for this type of task due to biases within the data, and poor real-world performance. Using autism spectrum disorder (ASD) as a test case, we sought to investigate the question: Can machine learning aid in the discovery of disease genes? We collected thirteen published ASD gene prioritization studies and evaluated their performance using known and novel high-confidence ASD genes. We also investigated their biases towards generic gene annotations, like number of association publications. We found that ML methods which do not incorporate genetics information have limited utility for prioritization of ASD risk genes. These studies perform at a comparable level to generic measures of likelihood for the involvement of genes in any condition, and do not out-perform genetic association studies. Future efforts to discover disease genes should be focused on developing and validating statistical models for genetic association, specifically for association between rare variants and disease, rather than developing complex machine learning methods using complex heterogeneous biological data with unknown reliability.

scholarly journals Mixed Logistic Regression in Genome-Wide Association Studies

2020 ◽  
Author(s):  
Jacqueline Milet ◽  
Hervé Perdry
Keyword(s):  
Logistic Regression ◽  
Linear Models ◽  
Association Studies ◽  
Score Test ◽  
R Package ◽  
Genome Wide Association ◽  
Supplementary Information ◽  
Genome Wide Association Studies ◽  
Mixed Linear Models ◽  
Genome Wide

AbstractMotivationMixed linear models (MLM) have been widely used to account for population structure in case-control genome-wide association studies, the status being analyzed as a quantitative phenotype. Chen et al. proved that this method is inappropriate and proposed a score test for the mixed logistic regression (MLR). However this test does not allow an estimation of the variants’ effects.ResultsWe propose two computationally efficient methods to estimate the variants’ effects. Their properties are evaluated on two simulations sets, and compared with other methods (MLM, logistic regression). MLR performs the best in all circumstances. The variants’ effects are well evaluated by our methods, with a moderate bias when the effect sizes are large. Additionally, we propose a stratified QQ-plot, enhancing the diagnosis of p-values inflation or deflation, when population strata are not clearly identified in the sample.AvailabilityAll methods are implemented in the R package milorGWAS available at https://github.com/genostats/[email protected] informationSupplementary data are available at Bioinformatics online.

scholarly journals Machine learning methods applied to genotyping data capture interactions between single nucleotide variants in late onset Alzheimer's disease

2021 ◽  
Author(s):  
Magdalena Arnal Segura ◽  
Dietmar Fernandez ◽  
Claudia Giambartolomei ◽  
Giorgio Bini ◽  
Eleftherios Samaras ◽  
...  
Keyword(s):  
Machine Learning ◽  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Late Onset ◽  
Hot Spot ◽  
Association Studies ◽  
Machine Learning Algorithms ◽  
Genome Wide Association Studies ◽  
Single Nucleotide Variants ◽  
Single Nucleotide

INTRODUCTION Genome-wide association studies (GWAS) in late onset Alzheimer's disease (LOAD) provide lists of individual genetic determinants. However, GWAS are not good at capturing the synergistic effects among multiple genetic variants and lack good specificity. METHODS We applied tree-based machine learning algorithms (MLs) to discriminate LOAD (> 700 individuals) and age-matched unaffected subjects using single nucleotide variants (SNVs) from AD studies, obtaining specific genomic profiles with the prioritized SNVs. RESULTS The MLs prioritized a set of SNVs located in close proximity genes PVRL2, TOMM40, APOE and APOC1. The captured genomic profiles in this region showed a clear interaction between rs405509 and rs1160985. Additionally, rs405509 located in APOE promoter interacts with rs429358 among others, seemingly neutralizing their predisposing effect. Interactions are characterized by their association with specific comorbidities and the presence of eQTL and sQTLs. DISCUSSION Our approach efficiently discriminates LOAD from controls, capturing genomic profiles defined by interactions among SNVs in a hot-spot region.

Sign in / Sign up

Export Citation Format

Share Document