1
|
Deek RA, Ma S, Lewis J, Li H. Statistical and computational methods for integrating microbiome, host genomics, and metabolomics data. eLife 2024; 13:e88956. [PMID: 38832759 PMCID: PMC11149933 DOI: 10.7554/elife.88956] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/26/2023] [Accepted: 05/10/2024] [Indexed: 06/05/2024] Open
Abstract
Large-scale microbiome studies are progressively utilizing multiomics designs, which include the collection of microbiome samples together with host genomics and metabolomics data. Despite the increasing number of data sources, there remains a bottleneck in understanding the relationships between different data modalities due to the limited number of statistical and computational methods for analyzing such data. Furthermore, little is known about the portability of general methods to the metagenomic setting and few specialized techniques have been developed. In this review, we summarize and implement some of the commonly used methods. We apply these methods to real data sets where shotgun metagenomic sequencing and metabolomics data are available for microbiome multiomics data integration analysis. We compare results across methods, highlight strengths and limitations of each, and discuss areas where statistical and computational innovation is needed.
Collapse
Affiliation(s)
- Rebecca A Deek
- Department of Biostatistics, University of PittsburghPittsburghUnited States
| | - Siyuan Ma
- Department of Biostatistics, Vanderbilt School of MedicineNashvilleUnited States
| | - James Lewis
- Division of Gastroenterology and Hepatology, Perelman School of Medicine, University of PennsylvaniaPhiladelphiaUnited States
| | - Hongzhe Li
- Department of Biostatistics, Epidemiology, and Informatics, Perelman School of Medicine, University of PennsylvaniaPhiladelphiaUnited States
| |
Collapse
|
2
|
Wang Y, Cheng T, Cui Y, Qu D, Peng X, Yang L, Xiao X. Associations between gut microbiota and adverse neurodevelopmental outcomes in preterm infants: a two-sample Mendelian randomization study. Front Neurosci 2024; 18:1344125. [PMID: 38419663 PMCID: PMC10899413 DOI: 10.3389/fnins.2024.1344125] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/25/2023] [Accepted: 02/05/2024] [Indexed: 03/02/2024] Open
Abstract
Gut microbiota are associated with adverse neurodevelopmental outcomes in preterm infants; however, the precise causal relationship remains unclear. In this study, we conducted a two-sample Mendelian randomization (MR) analysis to comprehensively study the relationship between gut microbiota and adverse neurodevelopmental outcomes in preterm infants and identify specific causal bacteria that may be associated with the occurrence and development of adverse neurodevelopmental outcomes in preterm infants. The genome-wide association analysis (GWAS) of the MiBioGen biogroup was used as the exposure data. The GWAS of six common adverse neurodevelopmental outcomes in premature infants from the FinnGen consortium R9 was used as the outcome data. Genetic variations, namely, single nucleotide polymorphisms (SNPs) below the locus-wide significance level (1 × 10-5) and genome-wide statistical significance threshold (5 × 10-8) were selected as instrumental variables (IVs). MR studies use inverse variance weighting (IVW) as the main method. To supplement this, we also applied three additional MR methods: MR-Egger, weighted median, and weighted mode. In addition, the Cochrane's Q test, MR-Egger intercept test, Mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO), and leave-one-out methods were used for sensitivity analysis. Our study shows a causal relationship between specific gut microbiota and neurodevelopmental outcomes in preterm infants. These findings provide new insights into the mechanism by which gut microbiota may mediate adverse neurodevelopmental outcomes in preterm infants.
Collapse
Affiliation(s)
- Yuqian Wang
- Department of Graduate, Dalian Medical University, Dalian, Liaoning, China
- Department of Pediatrics, The Second Hospital of Dalian Medical University, Dalian, Liaoning, China
| | - Tongfei Cheng
- Department of Pediatrics, The Affiliated Women’s and Children’s Hospital of Qingdao University, Qingdao, Shandong, China
| | - Yifan Cui
- Department of Pediatrics, Dalian Women and Children’s Medical Group, Dalian, Liaoning, China
| | - Danyang Qu
- Department of Pediatrics, The Second Hospital of Dalian Medical University, Dalian, Liaoning, China
| | - Xin Peng
- Department of Pediatrics, The Second Hospital of Dalian Medical University, Dalian, Liaoning, China
| | - Liu Yang
- Department of Pediatrics, The Second Hospital of Dalian Medical University, Dalian, Liaoning, China
| | - Xuwu Xiao
- Department of Graduate, Dalian Medical University, Dalian, Liaoning, China
- Department of Pediatrics, Dalian Women and Children’s Medical Group, Dalian, Liaoning, China
| |
Collapse
|
3
|
Beghetti I, Barone M, Brigidi P, Sansavini A, Corvaglia L, Aceti A, Turroni S. Early-life gut microbiota and neurodevelopment in preterm infants: a narrative review. Front Nutr 2023; 10:1241303. [PMID: 37614746 PMCID: PMC10443645 DOI: 10.3389/fnut.2023.1241303] [Citation(s) in RCA: 4] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/16/2023] [Accepted: 07/27/2023] [Indexed: 08/25/2023] Open
Abstract
Infants born preterm are at a high risk of both gut microbiota (GM) dysbiosis and neurodevelopmental impairment. While the link between early dysbiosis and short-term clinical outcomes is well established, the relationship with long-term infant health has only recently gained interest. Notably, there is a significant overlap in the developmental windows of GM and the nervous system in early life. The connection between GM and neurodevelopment was first described in animal models, but over the last decade a growing body of research has also identified GM features as one of the potential mediators for human neurodevelopmental and neuropsychiatric disorders. In this narrative review, we provide an overview of the developing GM in early life and its prospective relationship with neurodevelopment, with a focus on preterm infants. Animal models have provided evidence for emerging pathways linking early-life GM with brain development. Furthermore, a relationship between both dynamic patterns and static features of the GM during preterm infants' early life and brain maturation, as well as neurodevelopmental outcomes in early childhood, was documented. Future human studies in larger cohorts, integrated with studies on animal models, may provide additional evidence and help to identify predictive biomarkers and potential therapeutic targets for healthy neurodevelopment in preterm infants.
Collapse
Affiliation(s)
- Isadora Beghetti
- Neonatal Intensive Care Unit, IRCCS Azienda Ospedaliero-Universitaria di Bologna, Bologna, Italy
- Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy
| | - Monica Barone
- Microbiomics Unit, Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy
| | - Patrizia Brigidi
- Microbiomics Unit, Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy
| | - Alessandra Sansavini
- Department of Psychology “Renzo Canestrari”, University of Bologna, Bologna, Italy
| | - Luigi Corvaglia
- Neonatal Intensive Care Unit, IRCCS Azienda Ospedaliero-Universitaria di Bologna, Bologna, Italy
- Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy
| | - Arianna Aceti
- Neonatal Intensive Care Unit, IRCCS Azienda Ospedaliero-Universitaria di Bologna, Bologna, Italy
- Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy
| | - Silvia Turroni
- Unit of Microbiome Science and Biotechnology, Department of Pharmacy and Biotechnology, University of Bologna, Bologna, Italy
| |
Collapse
|
4
|
Li G, Li Y, Chen K. It's all relative: Regression analysis with compositional predictors. Biometrics 2023; 79:1318-1329. [PMID: 35616500 PMCID: PMC9767704 DOI: 10.1111/biom.13703] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/16/2021] [Accepted: 05/18/2022] [Indexed: 01/05/2023]
Abstract
Compositional data reside in a simplex and measure fractions or proportions of parts to a whole. Most existing regression methods for such data rely on log-ratio transformations that are inadequate or inappropriate in modeling high-dimensional data with excessive zeros and hierarchical structures. Moreover, such models usually lack a straightforward interpretation due to the interrelation between parts of a composition. We develop a novel relative-shift regression framework that directly uses proportions as predictors. The new framework provides a paradigm shift for regression analysis with compositional predictors and offers a superior interpretation of how shifting concentration between parts affects the response. New equi-sparsity and tree-guided regularization methods and an efficient smoothing proximal gradient algorithm are developed to facilitate feature aggregation and dimension reduction in regression. A unified finite-sample prediction error bound is derived for the proposed regularized estimators. We demonstrate the efficacy of the proposed methods in extensive simulation studies and a real gut microbiome study. Guided by the taxonomy of the microbiome data, the framework identifies important taxa at different taxonomic levels associated with the neurodevelopment of preterm infants.
Collapse
Affiliation(s)
- Gen Li
- Department of Biostatistics, School of Public Health, University of Michigan, Ann Arbor., Michigan, USA
| | - Yan Li
- Department of Biostatistics, School of Public Health, University of Michigan, Ann Arbor., Michigan, USA
| | - Kun Chen
- Department of Statistics, University of Connecticut, Connecticut, USA
| |
Collapse
|
5
|
Chen J, Li H, Zhao T, Chen K, Chen MH, Sun Z, Xu W, Maas K, Lester BM, Cong XS. The Impact of Early Life Experiences and Gut Microbiota on Neurobehavioral Development in Preterm Infants: A Longitudinal Cohort Study. Microorganisms 2023; 11:microorganisms11030814. [PMID: 36985387 PMCID: PMC10056840 DOI: 10.3390/microorganisms11030814] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/14/2023] [Revised: 03/11/2023] [Accepted: 03/16/2023] [Indexed: 03/30/2023] Open
Abstract
OBJECTIVES The objective of this study is to investigate the impact of early life experiences and gut microbiota on neurobehavioral development in preterm infants during neonatal intensive care unit (NICU) hospitalization. METHODS Preterm infants were followed from NICU admission until their 28th postnatal day or until discharge. Daily stool samples, painful/stressful experiences, feeding patterns, and other clinical and demographic data were collected. Gut microbiota was profiled using 16S rRNA sequencing, and operational taxonomic units (OTUs) were selected to predict the neurobehaviors. The neurobehavioral development was assessed by the Neonatal Neurobehavioral Scale (NNNS) at 36 to 38 weeks of post-menstrual age (PMA). Fifty-five infants who had NNNS measurements were included in the sparse log-contrast regression analysis. RESULTS Preterm infants who experienced a high level of pain/stress during the NICU hospitalization had higher NNNS stress/abstinence scores. Eight operational taxonomic units (OTUs) were identified to be associated with NNNS subscales after controlling demographic and clinical features, feeding patterns, and painful/stressful experiences. These OTUs and taxa belonging to seven genera, i.e., Enterobacteriaceae_unclassified, Escherichia-Shigella, Incertae_Sedis, Veillonella, Enterococcus, Clostridium_sensu_stricto_1, and Streptococcus with five belonging to Firmicutes and two belonging to Proteobacteria phylum. The enriched abundance of Enterobacteriaceae_unclassified (OTU17) and Streptococcus (OTU28) were consistently associated with less optimal neurobehavioral outcomes. The other six OTUs were also associated with infant neurobehavioral responses depending on days at NICU stay. CONCLUSIONS This study explored the dynamic impact of specific OTUs on neurobehavioral development in preterm infants after controlling for early life experiences, i.e., acute and chronic pain/stress and feeding in the NICU. The gut microbiota and acute pain/stressful experiences dynamically impact the neurobehavioral development in preterm infants during their NICU hospitalization.
Collapse
Affiliation(s)
- Jie Chen
- College of Nursing, Florida State University, Tallahassee, FL 32306, USA
- School of Nursing, University of Connecticut, Storrs, CT 06269, USA
| | - Hongfei Li
- Department of Statistics, University of Connecticut, Storrs, CT 06269, USA
| | - Tingting Zhao
- School of Nursing, University of Connecticut, Storrs, CT 06269, USA
- School of Nursing, Yale University, Orange, CT 06477, USA
| | - Kun Chen
- Department of Statistics, University of Connecticut, Storrs, CT 06269, USA
| | - Ming-Hui Chen
- Department of Statistics, University of Connecticut, Storrs, CT 06269, USA
| | - Zhe Sun
- Department of Statistics, University of Connecticut, Storrs, CT 06269, USA
- Department of Biostatistics, School of Public Health, Yale University, New Haven, CT 06520, USA
| | - Wanli Xu
- School of Nursing, University of Connecticut, Storrs, CT 06269, USA
| | - Kendra Maas
- Microbial Analysis, Resources, and Services (MARS), University of Connecticut, Storrs, CT 06269, USA
| | - Barry M Lester
- Brown Center for the Study of Children at Risk, Departments of Psychiatry and Pediatrics, Warren Alpert Medical School of Brown University, Providence, RI 02903, USA
| | - Xiaomei S Cong
- School of Nursing, University of Connecticut, Storrs, CT 06269, USA
- School of Nursing, Yale University, Orange, CT 06477, USA
- Institute for Systems Genomics, University of Connecticut, Farmington, CT 06030, USA
| |
Collapse
|
6
|
Chen J, Li H, Zhao T, Chen K, Chen MH, Sun Z, Xu W, Maas K, Lester B, Cong X. The impact of early life experiences and gut microbiota on neurobehavioral development among preterm infants: A longitudinal cohort study. MEDRXIV : THE PREPRINT SERVER FOR HEALTH SCIENCES 2023:2023.01.04.23284200. [PMID: 36711616 PMCID: PMC9882379 DOI: 10.1101/2023.01.04.23284200] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Indexed: 01/09/2023]
Abstract
Objectives To investigate the impact of early life experiences and gut microbiota on neurobehavioral development among preterm infants during neonatal intensive care unit (NICU) hospitalization. Methods Preterm infants were followed from the NICU admission until their 28 th postnatal day or until discharge. Daily stool samples, painful/stressful experiences, feeding patterns, and other clinical and demographic data were collected. Gut microbiota was profiled using 16S rRNA sequencing, and operational taxonomic units (OTUs) were selected to predict the neurobehaviors. The neurobehavioral development was assessed by the Neonatal Neurobehavioral Scale (NNNS) at 36 to 38 weeks of post-menstrual age (PMA). Fifty-five infants who had NNNS measurements were included in the sparse log-contrast regression analysis. Results Preterm infants who experienced high level of pain/stress during the NICU hospitalization that were associated with higher NNNS stress/abstinence scores. Eight operational taxonomic units (OTUs) were identified to be associated with of NNNS subscales after controlling demographic and clinical features, feeding patterns, and painful/stressful experiences. These OTUs, taxa belong to seven genera including Enterobacteriaceae_unclassified, Escherichia-Shigella, Incertae_Sedis, Veillonella, Enterococcus, Clostridium_sensu_stricto_1 , and Streptococcus with five belonging to Firmicutes and two belonging to Proteobacteria phylum. The enriched abundance of Enterobacteriaceae_unclassified (OTU17) and Streptococcus (OTU28) were consistently associated with less optimal neurobehavioral outcomes. The other six OTUs were also associated with infant neurobehavioral responses depending on days at NICU stay. Conclusions This study explored the dynamic impact of specific OTUs on neurobehavioral development among preterm infants after controlling for early life experiences, i.e., acute and chronic pain/stress, and feeding in the NICU.
Collapse
Affiliation(s)
- Jie Chen
- Florida State University College of Nursing, Tallahassee, FL., United States
- School of Nursing, University of Connecticut, Storrs, CT., United States
| | - Hongfei Li
- Department of Statistics, University of Connecticut, Storrs, CT., United States
| | - Tingting Zhao
- School of Nursing, University of Connecticut, Storrs, CT., United States
| | - Kun Chen
- Department of Statistics, University of Connecticut, Storrs, CT., United States
| | - Ming-Hui Chen
- Department of Statistics, University of Connecticut, Storrs, CT., United States
| | - Zhe Sun
- Department of Statistics, University of Connecticut, Storrs, CT., United States
- Department of Biostatistics, Yale School of Public Health, New Haven, CT., United States
| | - Wanli Xu
- School of Nursing, University of Connecticut, Storrs, CT., United States
| | - Kendra Maas
- University of Connecticut, Microbial Analysis, Resources, and Services (MARS), Storrs, CT., United States
| | - Barry Lester
- Brown Center for the Study of Children at Risk, Departments of Psychiatry and Pediatrics, Warren Alpert Medical School of Brown University, Providence, RI., United States
| | - Xiaomei Cong
- School of Nursing, University of Connecticut, Storrs, CT., United States
- Yale University School of Nursing, Orange, CT., United States
- Institute for Systems Genomics, University of Connecticut, Farmington, CT., United States
| |
Collapse
|
7
|
Vaher K, Bogaert D, Richardson H, Boardman JP. Microbiome-gut-brain axis in brain development, cognition and behavior during infancy and early childhood. DEVELOPMENTAL REVIEW 2022. [DOI: 10.1016/j.dr.2022.101038] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/15/2023]
|
8
|
Trends of fecal calprotectin levels and associations with early life experience in preterm infants. INTERDISCIPLINARY NURSING RESEARCH 2022; 1:36-42. [PMID: 36590866 PMCID: PMC9766919 DOI: 10.1097/nr9.0000000000000006] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 07/22/2022] [Accepted: 09/23/2022] [Indexed: 12/03/2022]
Abstract
Background Preterm infants are at risk for severe infections due to their immature immune systems. Factors such as early life pain/stress experiences and feeding may influence immune activation and maturation of immune systems. However, the underlying mechanism remains unclear. Fecal calprotectin (FCP) is a noninvasive surrogate biomarker of mucosal inflammation in the gastrointestinal tract and has been used in detecting intestinal inflammation in specific pediatric gastrointestinal disorders. Objective To describe the longitudinal trajectory of FCP levels in preterm infants and investigate the contributing factors that are associated with FCP levels. Design A longitudinal study design was used. Settings Preterm infants were recruited from 2 neonatal intensive care units (NICU) of a children's medical center in the North-eastern US. Methods Preterm infants were followed during their first 4 weeks of NICU hospitalization. Stool samples were collected twice per week to quantify the FCP levels. Cumulative pain/stress experiences and feeding types were measured daily. A linear mixed-effect model was used to examine the associations between FCP levels and demographic and clinical characteristics, cumulative pain/stress, and feeding over time. Results Forty-nine preterm infants were included in the study. Infants' FCP levels varied largely with a mean of 268.7±261.3 µg/g and increased over time. Preterm infants experienced an average of 7.5±5.0 acute painful procedures and 15.3±20.8 hours of chronic painful procedures per day during their NICU stay. The mean percentage of mother's own milk increased from the first week (57.1±36.5%) to the fourth week (60.7±38.9%) after birth. Elevated FCP concentration was associated with acute and cumulative (chronic) pain/stress levels, mother's own milk, non-White race, and higher severity of illness score. Conclusions FCP levels were elevated in preterm infants with wide interindividual and intraindividual variations. Cumulative pain/stress during the NICU hospitalization, feeding, race, and health status may influence FCP concentrations in early life that may be associated with inflammatory gut processes.
Collapse
|
9
|
|
10
|
Alenazi A. A review of compositional data analysis and recent advances. COMMUN STAT-THEOR M 2021. [DOI: 10.1080/03610926.2021.2014890] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/19/2022]
Affiliation(s)
- Abdulaziz Alenazi
- Department of Mathematics, College of Science, Northern Border University, Arar, Saudi Arabia
| |
Collapse
|
11
|
Liu X, Cong X, Li G, Maas K, Chen K. Multivariate log-contrast regression with sub-compositional predictors: Testing the association between preterm infants' gut microbiome and neurobehavioral outcomes. Stat Med 2021; 41:580-594. [PMID: 34897772 DOI: 10.1002/sim.9273] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/29/2020] [Revised: 09/25/2021] [Accepted: 11/15/2021] [Indexed: 11/10/2022]
Abstract
To link a clinical outcome with compositional predictors in microbiome analysis, the linear log-contrast model is a popular choice, and the inference procedure for assessing the significance of each covariate is also available. However, with the existence of multiple potentially interrelated outcomes and the information of the taxonomic hierarchy of bacteria, a multivariate analysis method that considers the group structure of compositional covariates and an accompanying group inference method are still lacking. Motivated by a study for identifying the microbes in the gut microbiome of preterm infants that impact their later neurobehavioral outcomes, we formulate a constrained integrative multi-view regression. The neurobehavioral scores form multivariate responses, the log-transformed sub-compositional microbiome data form multi-view feature matrices, and a set of linear constraints on their corresponding sub-coefficient matrices ensures the sub-compositional nature. We assume all the sub-coefficient matrices are possible of low-rank to enable joint selection and inference of sub-compositions/views. We propose a scaled composite nuclear norm penalization approach for model estimation and develop a hypothesis testing procedure through de-biasing to assess the significance of different views. Simulation studies confirm the effectiveness of the proposed procedure. We apply the method to the preterm infant study, and the identified microbes are mostly consistent with existing studies and biological understandings.
Collapse
Affiliation(s)
- Xiaokang Liu
- Department of Biostatistics, Epidemiology and Informatics, University of Pennsylvania, Philadelphia, Pennsylvania, USA
| | - Xiaomei Cong
- School of Nursing, University of Connecticut, Storrs, Connecticut, USA
| | - Gen Li
- Department of Biostatistics, University of Michigan, Ann Arbor, Michigan, USA
| | - Kendra Maas
- Microbial Analysis, Resources, and Services, University of Connecticut, Storrs, Connecticut, USA
| | - Kun Chen
- Department of Statistics, University of Connecticut, Storrs, Connecticut, USA
| |
Collapse
|
12
|
Sun Z, Xu W, Cong X, Li G, Chen K. LOG-CONTRAST REGRESSION WITH FUNCTIONAL COMPOSITIONAL PREDICTORS: LINKING PRETERM INFANT'S GUT MICROBIOME TRAJECTORIES TO NEUROBEHAVIORAL OUTCOME. Ann Appl Stat 2020; 14:1535-1556. [PMID: 34163544 DOI: 10.1214/20-aoas1357] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/19/2022]
Abstract
The neonatal intensive care unit (NICU) experience is known to be one of the most crucial factors that drive preterm infant's neurodevelopmental and health outcome. It is hypothesized that stressful early life experience of very preterm neonate is imprinting gut microbiome by the regulation of the so-called brain-gut axis, and consequently, certain microbiome markers are predictive of later infant neurodevelopment. To investigate, a preterm infant study was conducted; infant fecal samples were collected during the infants' first month of postnatal age, resulting in functional compositional microbiome data, and neurobehavioral outcomes were measured when infants reached 36-38 weeks of post-menstrual age. To identify potential microbiome markers and estimate how the trajectories of gut microbiome compositions during early postnatal stage impact later neurobehavioral outcomes of the preterm infants, we innovate a sparse log-contrast regression with functional compositional predictors. The functional simplex structure is strictly preserved, and the functional compositional predictors are allowed to have sparse, smoothly varying, and accumulating effects on the outcome through time. Through a pragmatic basis expansion step, the problem boils down to a linearly constrained sparse group regression, for which we develop an efficient algorithm and obtain theoretical performance guarantees. Our approach yields insightful results in the preterm infant study. The identified microbiome markers and the estimated time dynamics of their impact on the neurobehavioral outcome shed lights on the linkage between stress accumulation in early postnatal stage and neurodevelpomental process of infants.
Collapse
|