Predictive Modeling for Metabolomics Data

Abstract

In recent years, mass spectrometry (MS)-based metabolomics has been extensively applied to characterize biochemical mechanisms, and study physiological processes and phenotypic changes associated with disease. Metabolomics has also been important for identifying biomarkers of interest suitable for clinical diagnosis. For the purpose of predictive modeling, in this chapter, we will review various supervised learning algorithms such as random forest (RF), support vector machine (SVM), and partial least squares-discriminant analysis (PLS-DA). In addition, we will also review feature selection methods for identifying the best combination of metabolites for an accurate predictive model. We conclude with best practices for reproducibility by including internal and external replication, reporting metrics to assess performance, and providing guidelines to avoid overfitting and to deal with imbalanced classes. An analysis of an example data will illustrate the use of different machine learning methods and performance metrics.

Appendices

TRIPOD Checklist for Predictive Modeling for Metabolomics Data

Section/topic	Item	Checklist item	Section
Title and abstract
Title	1	Identify the study as developing and/or validating a multivariable prediction model, the target population, and the outcome to be predicted	See title
Abstract	2	Provide a summary of objectives, study design, setting, participants, sample size, predictors, outcome, statistical analysis, results, and conclusions	See abstract
Introduction
Background and objectives	3a	Explain the medical context (including whether diagnostic or prognostic) and rationale for developing or validating the multivariable prediction model, including references to existing models	Subheading 4.1
	3b	Specify the objectives, including whether the study describes the development or validation of the model or both	Internal validation, Subheading 4.4
Methods
Source of data	4a	Describe the study design or source of data (e.g., randomized trial, cohort, or registry data), separately for the development and validation data sets, if applicable	Subheading 4.1
	4b	Specify the key study dates, including start of accrual; end of accrual; and, if applicable, end of follow-up	Subheading 4.1, see [87]
Participants	5a	Specify key elements of the study setting (e.g., primary care, secondary care, general population) including number and location of centers	N/A
	5b	Describe eligibility criteria for participants	Subheading 4.1, see [87]
	5c	Give details of treatments received, if relevant	N/A
Outcome	6a	Clearly define the outcome that is predicted by the prediction model, including how and when assessed	Subheading 4.1
	6b	Report any actions to blind assessment of the outcome to be predicted	N/A
Predictors	7a	Clearly define all predictors used in developing or validating the multivariable prediction model, including how and when they were measured	2999 predictors, for more details see [87]
	7b	Report any actions to blind assessment of predictors for the outcome and other predictors	N/A
Sample size	8	Explain how the study size was arrived at	Subheading 4.1, see [87]
Missing data	9	Describe how missing data were handled (e.g., complete-case analysis, single imputation, multiple imputation) with details of any imputation method	The data was already preprocessed and imputed, see Subheading 4.1
Statistical analysis methods	10c	For validation, describe how the predictions were calculated	Subheading 3.3
	10d	Specify all measures used to assess model performance and, if relevant, to compare multiple models	Subheading 3.3
	10e	Describe any model updating (e.g., recalibration) arising from the validation, if done	N/A
Risk groups	11	Provide details on how risk groups were created, if done	N/A
Development vs. validation	12	For validation, identify any differences from the development data in setting, eligibility criteria, outcome, and predictors	N/A
Results
Participants	13a	Describe the flow of participants through the study, including the number of participants with and without the outcome and, if applicable, a summary of the follow-up time. A diagram may be helpful	Subheading 4.1, see [87]
	13b	Describe the characteristics of the participants (basic demographics, clinical features, available predictors), including the number of participants with missing data for predictors and outcome	Subheading 4.1, see [87]
	13c	For validation, show a comparison with the development data of the distribution of important variables (demographics, predictors, and outcome)	Subheadings 4.3 and 4.4
Model performance	16	Report performance measures (with CIs) for the prediction model	N/A
Model-updating	17	If done, report the results from any model updating (i.e., model specification, model performance)	Subheading 4.4
Discussion
Limitations	18	Discuss any limitations of the study (such as nonrepresentative sample, few events per predictor, missing data)	Subheading 4.1, see [87]
Interpretation	19a	For validation, discuss the results with reference to performance in the development data, and any other validation data	N/A
	19b	Give an overall interpretation of the results, considering objectives, limitations, results from similar studies, and other relevant evidence	Subheadings 4.4 and 5
Implications	20	Discuss the potential clinical use of the model and implications for future research	Subheadings 4.4 and 5. However, performance of the model is data-driven
Other information
Supplementary information	21	Provide information about the availability of supplementary resources, such as study protocol, web calculator, and data sets	Subheading 4.1, see [87]
Funding	22	Give the source of funding and the role of the funders for the present study	NIH

Selected Open Source (R/Bioconductor/Web-Based) Tools for Supervised Learning Algorithms

Method	Source	Reference
PLS-DA	Bioconductor (ropls)	[93]
PLS-DA, RF, and SVM	Bioconductor (biosigner)	[94]
SVM, RF	Bioconductor (MLSeq)	[95]
RF, SVM, PLS-DA	Metaboanalyst http://www.metaboanalyst.ca/	[96]
PCA, PLS-DA, RF	Bioconductor (statTarget)	[97]
Feature selection, metric evaluation	Bioconductor (OmicsMarker)	[98]
Sparse PLS-DA	Bioconductor (mixOmics)	[99]
Feature selection, metric evaluation	CRAN (lilikoi)	[100]
Probabilistic principal component analysis	CRAN (MetabolAnalyze)	[101]
Kernel-based metabolite differential analysis	CRAN (KMDA)	[21]
PLS-DA, OPLS-DA	CRAN (muma)	[102]
RF	CRAN (RFmarkerDetector)	[103]
RF, SVM, PLS-DA	CRAN (caret)	[104]