Embedding covariate adjustments in tree-based automated machine learning for biomedical big data analyses

Blood-based DNA methylation and exposure risk scores predict PTSD with high accuracy in military and civilian cohorts

Background

Incorporating genomic data into risk prediction has become an increasingly useful approach for rapid identification of individuals most at risk for complex disorders such as PTSD. Our goal was to develop and validate Methylation Risk Scores (MRS) using machine learning to distinguish individuals who have PTSD from those who do not.

Methods

Elastic Net was used to develop three risk score models using a discovery dataset (n = 1226; 314 cases, 912 controls) comprised of 5 diverse cohorts with available blood-derived DNA methylation (DNAm) measured on the Illumina Epic BeadChip. The first risk score, exposure and methylation risk score (eMRS) used cumulative and childhood trauma exposure and DNAm variables; the second, methylation-only risk score (MoRS) was based solely on DNAm data; the third, methylation-only risk scores with adjusted exposure variables (MoRSAE) utilized DNAm data adjusted for the two exposure variables. The potential of these risk scores to predict future PTSD based on pre-deployment data was also assessed. External validation of risk scores was conducted in four independent cohorts.

Results

The eMRS model showed the highest accuracy (92%), precision (91%), recall (87%), and f1-score (89%) in classifying PTSD using 3730 features. While still highly accurate, the MoRS (accuracy = 89%) using 3728 features and MoRSAE (accuracy = 84%) using 4150 features showed a decline in classification power. eMRS significantly predicted PTSD in one of the four independent cohorts, the BEAR cohort (beta = 0.6839, p-0.003), but not in the remaining three cohorts. Pre-deployment risk scores from all models (eMRS, beta = 1.92; MoRS, beta = 1.99 and MoRSAE, beta = 1.77) displayed a significant (p < 0.001) predictive power for post-deployment PTSD.

Conclusion

Results, especially those from the eMRS, reinforce earlier findings that methylation and trauma are interconnected and can be leveraged to increase the correct classification of those with vs. without PTSD. Moreover, our models can potentially be a valuable tool in predicting the future risk of developing PTSD. As more data become available, including additional molecular, environmental, and psychosocial factors in these scores may enhance their accuracy in predicting the condition and, relatedly, improve their performance in independent cohorts.

MiTree: A Unified Web Cloud Analytic Platform for User-Friendly and Interpretable Microbiome Data Mining Using Tree-Based Methods

The advent of next-generation sequencing has greatly accelerated the field of human microbiome studies. Currently, investigators are seeking, struggling and competing to find new ways to diagnose, treat and prevent human diseases through the human microbiome. Machine learning is a promising approach to help such an effort, especially due to the high complexity of microbiome data. However, many of the current machine learning algorithms are in a "black box", i.e., they are difficult to understand and interpret. In addition, clinicians, public health practitioners and biologists are not usually skilled at computer programming, and they do not always have high-end computing devices. Thus, in this study, we introduce a unified web cloud analytic platform, named MiTree, for user-friendly and interpretable microbiome data mining. MiTree employs tree-based learning methods, including decision tree, random forest and gradient boosting, that are well understood and suited to human microbiome studies. We also stress that MiTree can address both classification and regression problems through covariate-adjusted or unadjusted analysis. MiTree should serve as an easy-to-use and interpretable data mining tool for microbiome-based disease prediction modeling, and should provide new insights into microbiome-based diagnostics, treatment and prevention. MiTree is an open-source software that is available on our web server.

A Data-Driven Analysis of Ward Capacity Strain Metrics That Predict Clinical Outcomes Among Survivors of Acute Respiratory Failure

Supply-demand mismatch of ward resources ("ward capacity strain") alters care and outcomes. Narrow strain definitions and heterogeneous populations limit strain literature. Evaluate the predictive utility of a large set of candidate strain variables for in-hospital mortality and discharge destination among acute respiratory failure (ARF) survivors. In a retrospective cohort of ARF survivors transferred from intensive care units (ICUs) to wards in five hospitals from 4/2017-12/2019, we applied 11 machine learning (ML) models to identify ward strain measures during the first 24 hours after transfer most predictive of outcomes. Measures spanned patient volume (census, admissions, discharges), staff workload (medications administered, off-ward transports, transfusions, isolation precautions, patients per respiratory therapist and nurse), and average patient acuity (Laboratory Acute Physiology Score version 2, ICU transfers) domains. The cohort included 5,052 visits in 43 wards. Median age was 65 years (IQR 56-73); 2,865 (57%) were male; and 2,865 (57%) were white. 770 (15%) patients died in the hospital or had hospice discharges, and 2,628 (61%) were discharged home and 964 (23%) to skilled nursing facilities (SNFs). Ward admissions, isolation precautions, and hospital admissions most consistently predicted in-hospital mortality across ML models. Patients per nurse most consistently predicted discharge to home and SNF, and medications administered predicted SNF discharge. In this hypothesis-generating analysis of candidate ward strain variables' prediction of outcomes among ARF survivors, several variables emerged as consistently predictive of key outcomes across ML models. These findings suggest targets for future inferential studies to elucidate mechanisms of ward strain's adverse effects.

Automated quantitative trait locus analysis (AutoQTL)

BACKGROUND

Quantitative Trait Locus (QTL) analysis and Genome-Wide Association Studies (GWAS) have the power to identify variants that capture significant levels of phenotypic variance in complex traits. However, effort and time are required to select the best methods and optimize parameters and pre-processing steps. Although machine learning approaches have been shown to greatly assist in optimization and data processing, applying them to QTL analysis and GWAS is challenging due to the complexity of large, heterogenous datasets. Here, we describe proof-of-concept for an automated machine learning approach, AutoQTL, with the ability to automate many complicated decisions related to analysis of complex traits and generate solutions to describe relationships that exist in genetic data.

RESULTS

Using a publicly available dataset of 18 putative QTL from a large-scale GWAS of body mass index in the laboratory rat, Rattus norvegicus, AutoQTL captures the phenotypic variance explained under a standard additive model. AutoQTL also detects evidence of non-additive effects including deviations from additivity and 2-way epistatic interactions in simulated data via multiple optimal solutions. Additionally, feature importance metrics provide different insights into the inheritance models and predictive power of multiple GWAS-derived putative QTL.

CONCLUSIONS

This proof-of-concept illustrates that automated machine learning techniques can complement standard approaches and have the potential to detect both additive and non-additive effects via various optimal solutions and feature importance metrics. In the future, we aim to expand AutoQTL to accommodate omics-level datasets with intelligent feature selection and feature engineering strategies.

Reducing the complexity of high-dimensional environmental data: An analytical framework using LASSO with considerations of confounding for statistical inference

PURPOSE

Frameworks for selecting exposures in high-dimensional environmental datasets, while considering confounding, are lacking. We present a two-step approach for exposure selection with subsequent confounder adjustment for statistical inference.

METHODS

We measured cognitive ability in 338 children using the Woodcock-Muñoz General Intellectual Ability (GIA) score, and potential associated features across several environmental domains. Initially, 111 variables theoretically associated with GIA score were introduced into a Least Absolute Shrinkage and Selection Operator (LASSO) in a 50% feature selection subsample. Effect estimates for selected features were subsequently modeled in linear regressions in a 50% inference (hold out) subsample, first adjusting for sex and age and later for covariates selected via directed acyclic graphs (DAGs). All models were adjusted for clustering by school.

RESULTS

Of the 15 LASSO selected variables, eleven were not associated with GIA score following our inference modeling approach. Four variables were associated with GIA scores, including: serum ferritin adjusted for inflammation (inversely), mother's IQ (positively), father's education (positively), and hours per day the child works on homework (positively). Serum ferritin was not in the expected direction.

CONCLUSIONS

Our two-step approach moves high-dimensional feature selection a step further by incorporating DAG-based confounder adjustment for statistical inference.

Automated quantitative trait locus analysis (AutoQTL)

Background

Quantitative Trait Locus (QTL) analysis and Genome-Wide Association Studies (GWAS) have the power to identify variants that capture significant levels of phenotypic variance in complex traits. However, effort and time are required to select the best methods and optimize parameters and pre-processing steps. Although machine learning approaches have been shown to greatly assist in optimization and data processing, applying them to QTL analysis and GWAS is challenging due to the complexity of large, heterogenous datasets. Here, we describe proof-of-concept for an automated machine learning approach, AutoQTL, with the ability to automate many complex decisions related to analysis of complex traits and generate diverse solutions to describe relationships that exist in genetic data.

Results

Using a dataset of 18 putative QTL from a large-scale GWAS of body mass index in the laboratory rat, Rattus norvegicus , AutoQTL captures the phenotypic variance explained under a standard additive model while also providing evidence of non-additive effects including deviations from additivity and 2-way epistatic interactions from simulated data via multiple optimal solutions. Additionally, feature importance metrics provide different insights into the inheritance models and predictive power of multiple GWAS-derived putative QTL.

Conclusions

This proof-of-concept illustrates that automated machine learning techniques can be applied to genetic data and has the potential to detect both additive and non-additive effects via various optimal solutions and feature importance metrics. In the future, we aim to expand AutoQTL to accommodate omics-level datasets with intelligent feature selection strategies.

Genetic Analysis of Coronary Artery Disease Using Tree-Based Automated Machine Learning Informed By Biology-Based Feature Selection

Machine Learning (ML) approaches are increasingly being used in biomedical applications. Important challenges of ML include choosing the right algorithm and tuning the parameters for optimal performance. Automated ML (AutoML) methods, such as Tree-based Pipeline Optimization Tool (TPOT), have been developed to take some of the guesswork out of ML thus making this technology available to users from more diverse backgrounds. The goals of this study were to assess applicability of TPOT to genomics and to identify combinations of single nucleotide polymorphisms (SNPs) associated with coronary artery disease (CAD), with a focus on genes with high likelihood of being good CAD drug targets. We leveraged public functional genomic resources to group SNPs into biologically meaningful sets to be selected by TPOT. We applied this strategy to data from the U.K. Biobank, detecting a strikingly recurrent signal stemming from a group of 28 SNPs. Importance analysis of these SNPs uncovered functional relevance of the top SNPs to genes whose association with CAD is supported in the literature and other resources. Furthermore, we employed game-theory based metrics to study SNP contributions to individual-level TPOT predictions and discover distinct clusters of well-predicted CAD cases. The latter indicates a promising approach towards precision medicine.

What makes a good prediction? Feature importance and beginning to open the black box of machine learning in genetics

Genetic data have become increasingly complex within the past decade, leading researchers to pursue increasingly complex questions, such as those involving epistatic interactions and protein prediction. Traditional methods are ill-suited to answer these questions, but machine learning (ML) techniques offer an alternative solution. ML algorithms are commonly used in genetics to predict or classify subjects, but some methods evaluate which features (variables) are responsible for creating a good prediction; this is called feature importance. This is critical in genetics, as researchers are often interested in which features (e.g., SNP genotype or environmental exposure) are responsible for a good prediction. This allows for the deeper analysis beyond simple prediction, including the determination of risk factors associated with a given phenotype. Feature importance further permits the researcher to peer inside the black box of many ML algorithms to see how they work and which features are critical in informing a good prediction. This review focuses on ML methods that provide feature importance metrics for the analysis of genetic data. Five major categories of ML algorithms: k nearest neighbors, artificial neural networks, deep learning, support vector machines, and random forests are described. The review ends with a discussion of how to choose the best machine for a data set. This review will be particularly useful for genetic researchers looking to use ML methods to answer questions beyond basic prediction and classification.