Early Stopping in Experimentation With Real-Time Functional Magnetic Resonance Imaging Using a Modified Sequential Probability Ratio Test.

Introduction: Functional magnetic resonance imaging (fMRI) often involves long scanning durations to ensure the associated brain activity can be detected. However, excessive experimentation can lead to many undesirable effects, such as from learning and/or fatigue effects, discomfort for the subject, excessive motion artifacts and loss of sustained attention on task. Overly long experimentation can thus have a detrimental effect on signal quality and accurate voxel activation detection. Here, we propose dynamic experimentation with real-time fMRI using a novel statistically driven approach that invokes early stopping when sufficient statistical evidence for assessing the task-related activation is observed. Methods: Voxel-level sequential probability ratio test (SPRT) statistics based on general linear models (GLMs) were implemented on fMRI scans of a mathematical 1-back task from 12 healthy teenage subjects and 11 teenage subjects born extremely preterm (EPT). This approach is based on likelihood ratios and allows for systematic early stopping based on target statistical error thresholds. We adopt a two-stage estimation approach that allows for accurate estimates of GLM parameters before stopping is considered. Early stopping performance is reported for different first stage lengths, and activation results are compared with full durations. Finally, group comparisons are conducted with both early stopped and full duration scan data. Numerical parallelization was employed to facilitate completion of computations involving a new scan within every repetition time (TR). Results: Use of SPRT demonstrates the feasibility and efficiency gains of automated early stopping, with comparable activation detection as with full protocols. Dynamic stopping of stimulus administration was achieved in around half of subjects, with typical time savings of up to 33% (4 min on a 12 min scan). A group analysis produced similar patterns of activity for control subjects between early stopping and full duration scans. The EPT group, individually, demonstrated more variability in location and extent of the activations compared to the normal term control group. This was apparent in the EPT group results, reflected by fewer and smaller clusters. Conclusion: A systematic statistical approach for early stopping with real-time fMRI experimentation has been implemented. This dynamic approach has promise for reducing subject burden and fatigue effects.

Hadoop and PySpark for reproducibility and scalability of genomic sequencing studies.

Modern genomic studies are rapidly growing in scale, and the analytical approaches used to analyze genomic data are increasing in complexity. Genomic data management poses logistic and computational challenges, and analyses are increasingly reliant on genomic annotation resources that create their own data management and versioning issues. As a result, genomic datasets are increasingly handled in ways that limit the rigor and reproducibility of many analyses. In this work, we examine the use of the Spark infrastructure for the management, access, and analysis of genomic data in comparison to traditional genomic workflows on typical cluster environments. We validate the framework by reproducing previously published results from the Alzheimer's Disease Sequencing Project. Using the framework and analyses designed using Jupyter notebooks, Spark provides improved workflows, reduces user-driven data partitioning, and enhances the portability and reproducibility of distributed analyses required for large-scale genomic studies.

Association between AVPR1A, DRD2, and ASPM and endophenotypes of communication disorders.

OBJECTIVES: Speech sound disorder (SSD) is one of the most common communication disorders, with a prevalence rate of 16% at 3 years of age, and an estimated 3.8% of children still presenting speech difficulties at 6 years of age. Several studies have identified promising associations between communication disorders and genes in brain and neuronal pathways; however, there have been few studies focusing on SSD and its associated endophenotypes. On the basis of the hypothesis that neuronal genes may influence endophenotypes common to communication disorders, we focused on three genes related to brain and central nervous system functioning: the dopamine D2 receptor (DRD2) gene, the arginine-vasopressin receptor 1a (AVPR1A) gene, and the microcephaly-associated protein gene (ASPM). METHODS: We examined the association of these genes with key endophenotypes of SSD - phonological memory measured through multisyllabic and nonword repetition, vocabulary measured using the Expressive One Word Picture Vocabulary Test and Peabody Picture Vocabulary Test, and reading decoding measured using the Woodcock Reading Mastery Tests Revised - as well as with the clinical phenotype of SSD. We genotyped tag single nucleotide polymorphisms in these genes and examined 498 individuals from 180 families. RESULTS: These data show that several single nucleotide polymorphisms in all three genes were associated with phonological memory, vocabulary, and reading decoding, with P less than 0.05. Notably, associations in AVPR1A (rs11832266) were significant after multiple testing correction. Gene-level tests showed that DRD2 was associated with vocabulary, ASPM with vocabulary and reading decoding, and AVPR1A with all three endophenotypes. CONCLUSION: Endophenotypes common to SSD, language impairment, and reading disability are all associated with these neuronal pathway genes.

Association of smoking and other risk factors with Fuchs' endothelial corneal dystrophy severity and corneal thickness.

PURPOSE: We investigated effects of smoking and other risk factors on the development of advanced Fuchs' endothelial corneal dystrophy (FECD) and on central corneal thickness (CCT). METHODS: Eyes from Caucasian probands, affected and unaffected family members, and unrelated controls matched for age from the FECD Genetics Multi-Center Study (n = 2044 subjects) were examined. Univariate and multivariate models, adjusted for family correlations, were used to determine the effect of smoking, sex, diabetes, and age on FECD case/control status and CCT. RESULTS: In a multivariate model, sex and smoking were associated significantly with advanced FECD (grades 4-6) development (P = 0.016 and P = 0.047, respectively). Female sex increased odds by 34%. Smoking increased odds by 30%. In a multivariate model, diabetes was associated with an increase of 9.1 µm in average CCT (P = 0.021). Female sex was associated significantly with a decrease in average CCT by 6.9 µm (P = 0.015). Smoking had no significant effect on CCT in any model. As shown previously, advanced FECD was associated with large increases in CCT (31.4-94.2 µm). CONCLUSIONS: Smoking was associated with an increased risk of advanced FECD and self-reported diabetes was associated with increased CCT. Further study of the impact of smoking and diabetes on FECD development and changes in corneal thickness is warranted.

Differing roles for TCF4 and COL8A2 in central corneal thickness and fuchs endothelial corneal dystrophy.

Fuchs endothelial corneal dystrophy (FECD) is the most common late-onset, vision-threatening corneal dystrophy in the United States, affecting about 4% of the population. Advanced FECD involves a thickening of the cornea from stromal edema and changes in Descemet membrane. To understand the relationship between FECD and central corneal thickness (CCT), we characterized common genetic variation in COL8A2 and TCF4, genes previously implicated in CCT and/or FECD. Other genes previously associated with FECD (PITX2, ZEB1, SLC4A11), and genes only known to affect CCT (COL5A1, FOXO1, AVGR8, ZNF469) were also interrogated. FECD probands, relatives and controls were recruited from 32 clinical sites; a total of 532 cases and 204 controls were genotyped and tested for association of FECD case/control status, a 7-step FECD severity scale and CCT, adjusting for age and sex. Association of FECD grade with TCF4 was highly significant (OR= 6.01 at rs613872; p = 4.8×10(-25)), and remained significant when adjusted for changes in CCT (OR= 4.84; p = 2.2×10(-16)). Association of CCT with TCF4 was also significant (p = 6.1×10(-7)), but was abolished with adjustment for FECD grade (p = 0.92). After adjusting for FECD grade, markers in other genes examined were modestly associated (p ∼ 0.001) with FECD and/or CCT. Thus, common variants in TCF4 appear to influence FECD directly, and CCT secondarily via FECD. Additionally, changes in corneal thickness due to the effect of other loci may modify disease severity, age-at-onset, or other biomechanical characteristics.

A multicenter study to map genes for Fuchs endothelial corneal dystrophy: baseline characteristics and heritability.

PURPOSE: To describe the methods for family and case-control recruitment for a multicenter genetic and associated heritability analyses of Fuchs endothelial corneal dystrophy (FECD). METHODS: Twenty-nine enrolling sites with 62 trained investigators and coordinators gathered individual and family information, graded the phenotype, and collected blood and/or saliva for genetic analysis on all individuals with and without FECD. The degree of FECD was assessed in a 0 to 6 semiquantitative scale using standardized clinical methods with pathological verification of FECD on at least 1 member of each family. Central corneal thickness was measured by ultrasonic pachymetry. RESULTS: Three hundred twenty-two families with 330 affected sibling pairs with FECD were enrolled and included a total of 650 sibling pairs of all disease grades. Using the entire 7-step FECD grading scale or a dichotomous definition of severe disease, heritability was assessed in families via sib-sib correlations. Both binary indicators of severe disease and semiquantitative measures of disease severity were significantly heritable, with heritability estimates of 30% for severe disease, 37% to 39% for FECD score, and 47% for central corneal thickness. CONCLUSIONS: Genetic risk factors have a strong role in the severity of the FECD phenotype and corneal thickness. Genotyping this cohort with high-density genetic markers followed by appropriate statistical analyses should lead to novel loci for disease susceptibility.