Gaussian and Mixed Graphical Models as (multi-)omics data analysis tools

Altenbuchinger, Michael; Weihs, Antoine; Quackenbush, John; Grabe, Hans J.; Zacharias, Helena U.

doi:10.1016/j.bbagrm.2019.194418

Cited by 47 publications

(40 citation statements)

References 82 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, the early strategy allows the inference of heterogeneous networks using methods such as Mixed Graphical Models (MGM) [75] , [76] , which expand from Gaussian Graphical Models that assume normal distribution of variables to a mixed model. MGM regresses each variable against every other using either linear regression or logistic regression depending on the type of variable (continuous or discrete/categorical).…”

Section: Main Integration Strategiesmentioning

confidence: 99%

“…Additionally, Zhong et al (2019) [80] developed mixed Directed Acyclic Graph (mDAG) which can infer causal interactions based on variables with different distributions and can potentially be used in multi-omics studies. More information on inferring heterogeneous networks from multi-omics data can be found in the reviews [76] , [81] .…”

Section: Main Integration Strategiesmentioning

confidence: 99%

See 1 more Smart Citation

Integration strategies of multi-omics data for machine learning analysis

Picard

Scott‐Boyer

Bodein

et al. 2021

Computational and Structural Biotechnology Journal

239

185

View full text Add to dashboard Cite

Graphical abstract Schematic representation of the main strategies for multi-omics datasets integration. A) Early integration concatenates all omics datasets into a single matrix on which machine learning model can be applied. B) Mixed integration first independently transforms or maps each omics block into a new representation before combining them for downstream analysis. C) Intermediate integration simultaneously transforms the original datasets into common and omics-specific representations. D) Late integration analyses each omics separately and combines their final predictions. E) Hierarchical integration bases the integration of datasets on prior regulatory relationships between omics layers.

show abstract

Section: Main Integration Strategiesmentioning

confidence: 99%

Section: Main Integration Strategiesmentioning

confidence: 99%

Integration strategies of multi-omics data for machine learning analysis

Picard

Scott‐Boyer

Bodein

et al. 2021

Computational and Structural Biotechnology Journal

239

185

View full text Add to dashboard Cite

show abstract

“…For this research, GGMs were designed and processed through the extended Bayesian information criteria (EBIC) with graphical (g) least absolute shrinkage and selection operator (lasso) (EBICglasso) and partial correlation (Pcor). A Gaussian graphical model (GGM) is a “graph in which all random variables are continuous and jointly Gaussian [ 11 , 12 ] and it is based on conditional independence, respectively if

, two variables v1 and v2 are conditionally independent if

, namely there are 0 entries of the precision matrix

. Σ is the positive definite covariance matrix and

is the precision matrix of the distribution, defined as the inverse of

”.…”

Section: Methodsmentioning

confidence: 99%

“…“If

is positive definite, distribution has density on

. The sample covariance matrix is represented by

” [ 11 , 12 , 13 ].…”

Section: Methodsmentioning

confidence: 99%

Individual Characteristics as Prognostic Factors of the Evolution of Hospitalized COVID-19 Romanian Patients: A Comparative Observational Study between the First and Second Waves Based on Gaussian Graphical Models and Structural Equation Modeling

Mocanu

Noja

Istodor

et al. 2021

JCM

View full text Add to dashboard Cite

This study examines the role played by individual characteristics and specific treatment methods in the evolution of hospitalized patients with coronavirus disease 2019 (COVID-19), through the lens of an observational study performed in a comparative approach between the first and second waves of coronavirus pandemic in Romania. The research endeavor is configured on a two-fold approach, including a detailed observation of the evolution of 274 hospitalized patients with COVID-19 (145 in the first wave and 129 in the second wave of infection) according to specific treatment methods applied and patients’ individual features, as well as an econometric (quantitative) analysis through structural equation modeling and Gaussian graphical models designed to acknowledge the correlations and causal relationship between all considered coordinates. The main results highlight that the specific treatment methods applied had a positive influence on the evolution of COVID-19 patients, particularly in the second wave of coronavirus pandemic. In case of the first wave of COVID-19 infection, GGM results entail that there is a strong positive correlation between the evolution of the patients and the COVID-19 disease form, which is further positively correlated with the treatment scheme. The evolution of the patients is strongly and inversely correlated with the symptomatology and the ICU hospitalization. Moreover, the disease form is strongly and inversely correlated with oxygen saturation and the residence of patients (urban/rural). The symptomatology at first appearance also strongly depends on the age of the patients (positive correlation) and of the fact that the patient is a smoker or non-smoker and has other comorbidities. Age and gender are also important credentials that shape the disease degree and patient evolution in responding to treatment as well, our study attesting strong interconnections between these coordinates, the form of disease, symptomatology and overall evolution of the patients.

show abstract

“…Additionally for the methylome, we checked whether CpGs located in the 4th quartile for intra-and interindividual variability were enriched for CpG island relative position (island, shore, shelf, open sea) and for overlap with CpGs associated with exposures/traits in the EWAS Atlas [53]. [35], CV coefficient of variation, BLD below limit of detection relevant associations [54][55][56][57][58]. GGMs were built on the delta matrix calculated as the change in omics features between visits (i.e.…”

Section: Linear Mixed Effect Modelsmentioning

confidence: 99%

Variability of multi-omics profiles in a population-based child cohort

Hernández-Ferrer

Bustamante

et al. 2021

BMC Med

View full text Add to dashboard Cite

Background Multiple omics technologies are increasingly applied to detect early, subtle molecular responses to environmental stressors for future disease risk prevention. However, there is an urgent need for further evaluation of stability and variability of omics profiles in healthy individuals, especially during childhood. Methods We aimed to estimate intra-, inter-individual and cohort variability of multi-omics profiles (blood DNA methylation, gene expression, miRNA, proteins and serum and urine metabolites) measured 6 months apart in 156 healthy children from five European countries. We further performed a multi-omics network analysis to establish clusters of co-varying omics features and assessed the contribution of key variables (including biological traits and sample collection parameters) to omics variability. Results All omics displayed a large range of intra- and inter-individual variability depending on each omics feature, although all presented a highest median intra-individual variability. DNA methylation was the most stable profile (median 37.6% inter-individual variability) while gene expression was the least stable (6.6%). Among the least stable features, we identified 1% cross-omics co-variation between CpGs and metabolites (e.g. glucose and CpGs related to obesity and type 2 diabetes). Explanatory variables, including age and body mass index (BMI), explained up to 9% of serum metabolite variability. Conclusions Methylation and targeted serum metabolomics are the most reliable omics to implement in single time-point measurements in large cross-sectional studies. In the case of metabolomics, sample collection and individual traits (e.g. BMI) are important parameters to control for improved comparability, at the study design or analysis stage. This study will be valuable for the design and interpretation of epidemiological studies that aim to link omics signatures to disease, environmental exposures, or both.

show abstract

Gaussian and Mixed Graphical Models as (multi-)omics data analysis tools

Cited by 47 publications

References 82 publications

Integration strategies of multi-omics data for machine learning analysis

Integration strategies of multi-omics data for machine learning analysis

Individual Characteristics as Prognostic Factors of the Evolution of Hospitalized COVID-19 Romanian Patients: A Comparative Observational Study between the First and Second Waves Based on Gaussian Graphical Models and Structural Equation Modeling

Variability of multi-omics profiles in a population-based child cohort

Contact Info

Product

Resources

About