Improving mass-univariate analysis of neuroimaging data by modelling important unknown covariates: Application to Epigenome-Wide Association Studies.

Statistical inference on neuroimaging data is often conducted using a mass-univariate model, equivalent to fitting a linear model at every voxel with a known set of covariates. Due to the large number of linear models, it is challenging to check if the selection of covariates is appropriate and to modify this selection adequately. The use of standard diagnostics, such as residual plotting, is clearly not practical for neuroimaging data. However, the selection of covariates is crucial for linear regression to ensure valid statistical inference. In particular, the mean model of regression needs to be reasonably well specified. Unfortunately, this issue is often overlooked in the field of neuroimaging. This study aims to adopt the existing Confounder Adjusted Testing and Estimation (CATE) approach and to extend it for use with neuroimaging data. We propose a modification of CATE that can yield valid statistical inferences using Principal Component Analysis (PCA) estimators instead of Maximum Likelihood (ML) estimators. We then propose a non-parametric hypothesis testing procedure that can improve upon parametric testing. Monte Carlo simulations show that the modification of CATE allows for more accurate modelling of neuroimaging data and can in turn yield a better control of False Positive Rate (FPR) and Family-Wise Error Rate (FWER). We demonstrate its application to an Epigenome-Wide Association Study (EWAS) on neonatal brain imaging and umbilical cord DNA methylation data obtained as part of a longitudinal cohort study. Software for this CATE study is freely available at http://www.bioeng.nus.edu.sg/cfa/Imaging_Genetics2.html.

The effect of genotype and in utero environment on interindividual variation in neonate DNA methylomes.

Integrating the genotype with epigenetic marks holds the promise of better understanding the biology that underlies the complex interactions of inherited and environmental components that define the developmental origins of a range of disorders. The quality of the in utero environment significantly influences health over the lifecourse. Epigenetics, and in particular DNA methylation marks, have been postulated as a mechanism for the enduring effects of the prenatal environment. Accordingly, neonate methylomes contain molecular memory of the individual in utero experience. However, interindividual variation in methylation can also be a consequence of DNA sequence polymorphisms that result in methylation quantitative trait loci (methQTLs) and, potentially, the interaction between fixed genetic variation and environmental influences. We surveyed the genotypes and DNA methylomes of 237 neonates and found 1423 punctuate regions of the methylome that were highly variable across individuals, termed variably methylated regions (VMRs), against a backdrop of homogeneity. MethQTLs were readily detected in neonatal methylomes, and genotype alone best explained ∼25% of the VMRs. We found that the best explanation for 75% of VMRs was the interaction of genotype with different in utero environments, including maternal smoking, maternal depression, maternal BMI, infant birth weight, gestational age, and birth order. Our study sheds new light on the complex relationship between biological inheritance as represented by genotype and individual prenatal experience and suggests the importance of considering both fixed genetic variation and environmental factors in interpreting epigenetic variation.

The influence of CHRNA4, COMT, and maternal sensitivity on orienting and executive attention in 6-month-old infants.

Despite claims concerning biological mechanisms sub-serving infant attention, little experimental work examines its underpinnings. This study examines how candidate polymorphisms from the cholinergic (CHRNA4 rs1044396) and dopaminergic (COMT rs4680) systems, respectively indicative of parietal and prefrontal/anterior cingulate involvement, are related to 6-month-olds' (n=217) performance during a visual expectation eye-tracking paradigm. As previous studies suggest that both cholinergic and dopaminergic genes may influence susceptibility to the influence of other genetic and environmental factors, we further examined whether these candidate genes interact with one another and/or with early caregiving experience in predicting infants' visual attention. We detected an interaction between CHRNA4 genotype and observed maternal sensitivity upon infants' orienting to random stimuli and a CHRNA4-COMT interaction effect upon infants' orienting to patterned stimuli. Consistent with adult research, we observed a direct effect of COMT genotype on anticipatory looking to patterned stimuli. Findings suggest that CHRNA4 genotype may influence susceptibility to other attention-related factors in infancy. These interactions may account for the inability to establish a link between CHRNA4 and orienting in infant research to date, despite developmental theorizing suggesting otherwise. Moreover, findings suggest that by 6months, dopamine, and relatedly, the prefrontal cortex/anterior cingulate, may be important to infant attention.

Neonatal amygdalae and hippocampi are influenced by genotype and prenatal environment, and reflected in the neonatal DNA methylome.

The amygdala and hippocampus undergo rapid development in early life. The relative contribution of genetic and environmental factors to the establishment of their developmental trajectories has yet to be examined. We performed imaging on neonates and examined how the observed variation in volume and microstructure of the amygdala and hippocampus varied by genotype, and compared with prenatal maternal mental health and socioeconomic status. Gene × Environment models outcompeted models containing genotype or environment only to best explain the majority of measures but some, especially of the amygdaloid microstructure, were best explained by genotype only. Models including DNA methylation measured in the neonate umbilical cords outcompeted the Gene and Gene × Environment models for the majority of amygdaloid measures and minority of hippocampal measures. This study identified brain region-specific gene networks associated with individual differences in fetal brain development. In particular, genetic and epigenetic variation within CUX1 was highlighted.

Predicting functional decline and survival in amyotrophic lateral sclerosis.

BACKGROUND: Better predictors of amyotrophic lateral sclerosis disease course could enable smaller and more targeted clinical trials. Partially to address this aim, the Prize for Life foundation collected de-identified records from amyotrophic lateral sclerosis sufferers who participated in clinical trials of investigational drugs and made them available to researchers in the PRO-ACT database. METHODS: In this study, time series data from PRO-ACT subjects were fitted to exponential models. Binary classes for decline in the total score of amyotrophic lateral sclerosis functional rating scale revised (ALSFRS-R) (fast/slow progression) and survival (high/low death risk) were derived. Data was segregated into training and test sets via cross validation. Learning algorithms were applied to the demographic, clinical and laboratory parameters in the training set to predict ALSFRS-R decline and the derived fast/slow progression and high/low death risk categories. The performance of predictive models was assessed by cross-validation in the test set using Receiver Operator Curves and root mean squared errors. RESULTS: A model created using a boosting algorithm containing the decline in four parameters (weight, alkaline phosphatase, albumin and creatine kinase) post baseline, was able to predict functional decline class (fast or slow) with fair accuracy (AUC = 0.82). However similar approaches to build a predictive model for decline class by baseline subject characteristics were not successful. In contrast, baseline values of total bilirubin, gamma glutamyltransferase, urine specific gravity and ALSFRS-R item score-climbing stairs were sufficient to predict survival class. CONCLUSIONS: Using combinations of small numbers of variables it was possible to predict classes of functional decline and survival across the 1-2 year timeframe available in PRO-ACT. These findings may have utility for design of future ALS clinical trials.

Viral genome imaging of hepatitis C virus to probe heterogeneous viral infection and responses to antiviral therapies.

Hepatitis C virus (HCV) is a positive single-stranded RNA virus of enormous global health importance, with direct-acting antiviral therapies replacing an immunostimulatory interferon-based regimen. The dynamics of HCV positive and negative-strand viral RNAs (vRNAs) under antiviral perturbations have not been studied at the single-cell level, leaving a gap in our understanding of antiviral kinetics and host-virus interactions. Here, we demonstrate quantitative imaging of HCV genomes in multiple infection models, and multiplexing of positive and negative strand vRNAs and host antiviral RNAs. We capture the varying kinetics with which antiviral drugs with different mechanisms of action clear HCV infection, finding the NS5A inhibitor daclatasvir to induce a rapid decline in negative-strand viral RNAs. We also find that the induction of host antiviral genes upon interferon treatment is positively correlated with viral load in single cells. This study adds smFISH to the toolbox available for analyzing the treatment of RNA virus infections.

COMT haplotypes modulate associations of antenatal maternal anxiety and neonatal cortical morphology.

OBJECTIVE: Exposure to antenatal maternal anxiety and complex genetic variations may shape fetal brain development. In particular, the catechol-O-methyltransferase (COMT) gene, located on chromosome 22q11.2, regulates catecholamine signaling in the prefrontal cortex and is implicated in anxiety, pain, and stress responsivity. This study examined whether individual single-nucleotide polymorphisms (SNPs) of the COMT gene and their haplotypes moderate the association between antenatal maternal anxiety and in utero cortical development. METHOD: A total of 146 neonates were genotyped and underwent MRI shortly after birth. Neonatal cortical morphology was characterized using cortical thickness. Antenatal maternal anxiety was assessed using the State-Trait Anxiety Inventory at week 26 of pregnancy. RESULTS: Individual COMT SNPs (val158met, rs737865, and rs165599) modulated the association between antenatal maternal anxiety and the prefrontal and parietal cortical thickness in neonates. Based on haplotype trend regression analysis, findings also showed that among rs737865-val158met-rs165599 haplotypes, the A-val-G (AGG) haplotype probabilities modulated positive associations of antenatal maternal anxiety with cortical thickness in the right ventrolateral prefrontal cortex and the right superior parietal cortex and precuneus. In contrast, the G-met-A (GAA) haplotype probabilities modulated negative associations of antenatal maternal anxiety with cortical thickness in bilateral precentral gyrus and the dorsolateral prefrontal cortex. CONCLUSIONS: These results suggest that the association between maternal anxiety and in utero neurodevelopment is modified through complex genetic variation in COMT. Such genetic moderation may explain, in part, the variation in phenotypic outcomes in offspring associated with maternal emotional well-being.

Novel region discovery method for Infinium 450K DNA methylation data reveals changes associated with aging in muscle and neuronal pathways.

We describe a methodology for detecting differentially methylated regions (DMRs) and variably methylated regions (VMRs), in data from Infinium 450K arrays that are very widely used in epigenetic studies. Region detection is more specific than single CpG analysis as it increases the extent of common findings between studies, and is more powerful as it reduces the multiple testing problem inherent in epigenetic whole-genome association studies (EWAS). In addition, results driven by single erroneous probes are removed. We have used multiple publicly available Infinium 450K data sets to generate a consensus list of DMRs for age, supporting the hypothesis that aging is associated with specific epigenetic modifications. The consensus aging DMRs are significantly enriched for muscle biogenesis pathways. We find a massive increase in VMRs with age and in regions of the genome associated with open chromatin and neurotransmission. Old age VMRs are significantly enriched for neurotransmission pathways. EWAS studies should investigate the role of this interindividual variation in DNA methylation, in the age-associated diseases of sarcopenia and dementia.

Infinium monkeys: Infinium 450K array for the Cynomolgus macaque (Macaca fascicularis).

The Infinium Human Methylation450 BeadChip Array (Infinium 450K) is a robust and cost-efficient survey of genome-wide DNA methylation patterns. Macaca fascicularis (Cynomolgus macaque) is an important disease model; however, its genome sequence is only recently published, and few tools exist to interrogate the molecular state of Cynomolgus macaque tissues. Although the Infinium 450K is a hybridization array designed to the human genome, the relative conservation between the macaque and human genomes makes its use in macaques feasible. Here, we used the Infinium 450K array to assay DNA methylation in 11 macaque muscle biopsies. We showed that probe hybridization efficiency was related to the degree of sequence identity between the human probes and the macaque genome sequence. Approximately 61% of the Human Infinium 450K probes could be reliably mapped to the Cynomolgus macaque genome and contain a CpG site of interest. We also compared the Infinium 450K data to reduced representation bisulfite sequencing data generated on the same samples and found a high level of concordance between the two independent methodologies, which can be further improved by filtering for probe sequence identity and mismatch location. We conclude that the Infinium 450K array can be used to measure the DNA methylome of Cynomolgus macaque tissues using the provided filters. We also provide a pipeline for validation of the array in other species using a simple BLAST-based sequence identify filter.

Measuring the methylome in clinical samples: improved processing of the Infinium Human Methylation450 BeadChip Array.

The Infinium Human Methylation450 BeadChip Array (TM) (Infinium 450K) is an important tool for studying epigenetic patterns associated with disease. This array offers a high-throughput, low cost alternative to more comprehensive sequencing-based methodologies. Here we compare data generated by interrogation of the same seven clinical samples by Infinium 450K and reduced representation bisulfite sequencing (RRBS). This is the largest data set comparing Infinium 450K array to the comprehensive RRBS methodology reported so far. We show good agreement between the two methodologies. A read depth of four or more reads in the RRBS data was sufficient to achieve good agreement with Infinium 450K. However, we observe that intermediate methylation values (20-80%) are more variable between technologies than values at the extremes of the bimodal methylation distribution. We describe careful processing of Infinium 450K data to correct for known limitations and batch effects. Using methodologies proposed by others and newly implemented and combined in this report, agreement of Infinium 450K data with independent techniques can be vastly improved.

Transcriptome changes affecting Hedgehog and cytokine signalling in the umbilical cord: implications for disease risk.

BACKGROUND: Babies born at lower gestational ages or smaller birthweights have a greater risk of poorer health in later life. Both the causes of these sub-optimal birth outcomes and the mechanism by which the effects are transmitted over decades are the subject of extensive study. We investigated whether a transcriptomic signature of either birthweight or gestational age could be detected in umbilical cord RNA. METHODS: The gene expression patterns of 32 umbilical cords from Singaporean babies of Chinese ethnicity across a range of birthweights (1698-4151 g) and gestational ages (35-41 weeks) were determined. We confirmed the differential expression pattern by gestational age for 12 genes in a series of 127 umbilical cords of Chinese, Malay and Indian ethnicity. RESULTS: We found that the transcriptome is substantially influenced by gestational age; but less so by birthweight. We show that some of the expression changes dependent on gestational age are enriched in signal transduction pathways, such as Hedgehog and in genes with roles in cytokine signalling and angiogenesis. We show that some of the gene expression changes we report are reflected in the epigenome. CONCLUSIONS: We studied the umbilical cord which is peripheral to disease susceptible tissues. The results suggest that soma-wide transcriptome changes, preserved at the epigenetic level, may be a mechanism whereby birth outcomes are linked to the risk of adult metabolic and arthritic disease and suggest that greater attention be given to the association between premature birth and later disease risk.