Jason M. Knight

Non-invasive analysis of intestinal development in preterm and term infants using RNA-Sequencing

The state and development of the intestinal epithelium is vital for infant health, and increased understanding in this area has been limited by an inability to directly assess epithelial cell biology in the healthy newborn intestine. To that end, we have developed a novel, noninvasive, molecular approach that utilizes next generation RNA sequencing on stool samples containing intact epithelial cells for the purpose of quantifying intestinal gene expression. We then applied this technique to compare host gene expression in healthy term and extremely preterm infants. Bioinformatic analyses demonstrate repeatable detection of human mRNA expression, and network analysis shows immune cell function and inflammation pathways to be up-regulated in preterm infants. This study provides incontrovertible evidence that whole-genome sequencing of stool-derived RNA can be used to examine the neonatal host epithelial transcriptome in infants, which opens up opportunities for sequential monitoring of gut gene expression in response to dietary or therapeutic interventions.

Comparative effects of diet and carcinogen on microRNA expression in the stem cell niche of the mouse colonic crypt

There is mounting evidence that noncoding microRNAs (miRNA) are modulated by select chemoprotective dietary agents. For example, recently we demonstrated that the unique combination of dietary fish oil (containing n-3 fatty acids) plus pectin (fermented to butyrate in the colon) (FPA) up-regulates a subset of putative tumor suppressor miRNAs in intestinal mucosa, and down-regulates their predicted target genes following carcinogen exposure as compared to control (corn oil plus cellulose (CCA)) diet. To further elucidate the biological effects of diet and carcinogen modulated miRs in the colon, we verified that miR-26b and miR-203 directly target PDE4B and TCF4, respectively. Since perturbations in adult stem cell dynamics are generally believed to represent an early step in colon tumorigenesis and to better understand how the colonic stem cell population responds to environmental factors such as diet and carcinogen, we additionally determined the effects of the chemoprotective FPA diet on miRNAs and mRNAs in colonic stem cells obtained from Lgr5-EGFP-IRES-creER(T2) knock-in mice. Following global miRNA profiling, 26 miRNAs (P<0.05) were differentially expressed in Lgr5(high) stem cells as compared to Lgr5(negative) differentiated cells. FPA treatment up-regulated miR-19b, miR-26b and miR-203 expression as compared to CCA specifically in Lgr5(high) cells. In contrast, in Lgr5(negative) cells, only miR-19b and its indirect target PTK2B were modulated by the FPA diet. These data indicate for the first time that select dietary cues can impact stem cell regulatory networks, in part, by modulating the steady-state levels of miRNAs. To our knowledge, this is the first study to utilize Lgr5(+) reporter mice to determine the impact of diet and carcinogen on miRNA expression in colonic stem cells and their progeny.

MCMC implementation of the optimal Bayesian classifier for non-Gaussian models: model-based RNA-Seq classification

BackgroundSequencing datasets consist of a finite number of reads which map to specific regions of a reference genome. Most effort in modeling these datasets focuses on the detection of univariate differentially expressed genes. However, for classification, we must consider multiple genes and their interactions.ResultsThus, we introduce a hierarchical multivariate Poisson model (MP) and the associated optimal Bayesian classifier (OBC) for classifying samples using sequencing data. Lacking closed-form solutions, we employ a Monte Carlo Markov Chain (MCMC) approach to perform classification. We demonstrate superior or equivalent classification performance compared to typical classifiers for two synthetic datasets and over a range of classification problem difficulties. We also introduce the Bayesian minimum mean squared error (MMSE) conditional error estimator and demonstrate its computation over the feature space. In addition, we demonstrate superior or leading class performance over an RNA-Seq dataset containing two lung cancer tumor types from The Cancer Genome Atlas (TCGA).ConclusionsThrough model-based, optimal Bayesian classification, we demonstrate superior classification performance for both synthetic and real RNA-Seq datasets. A tutorial video and Python source code is available under an open source license at http://bit.ly/1gimnss.

Classifier design given an uncertainty class of feature distributions via regularized maximum likelihood and the incorporation of biological pathway knowledge in steady-state phenotype classification

Contemporary high-throughput technologies provide measurements of very large numbers of variables but often with very small sample sizes. This paper proposes an optimization-based paradigm for utilizing prior knowledge to design better performing classifiers when sample sizes are limited. We derive approximate expressions for the first and second moments of the true error rate of the proposed classifier under the assumption of two widely-used models for the uncertainty classes; ε-contamination and p-point classes. The applicability of the approximate expressions is discussed by defining the problem of finding optimal regularization parameters through minimizing the expected true error. Simulation results using the Zipf model show that the proposed paradigm yields improved classifiers that outperform traditional classifiers that use only training data. Our application of interest involves discrete gene regulatory networks possessing labeled steady-state distributions. Given prior operational knowledge of the process, our goal is to build a classifier that can accurately label future observations obtained in the steady state by utilizing both the available prior knowledge and the training data. We examine the proposed paradigm on networks containing NF-κB pathways, where it shows significant improvement in classifier performance over the classical data-only approach to classifier design. Companion website: http://gsp.tamu.edu/Publications/supplementary/shahrokh12a.

Generating Stochastic Gene Regulatory Networks Consistent With Pathway Information and Steady-State Behavior

We present a procedure to generate a stochastic genetic regulatory network model consistent with pathway information. Using the stochastic dynamics of Markov chains, we produce a model constrained by the prior knowledge despite the sometimes incomplete, time independent, and often conflicting nature of these pathways. We apply the Markov theory to study the models long run behavior and introduce a biologically important transformation to aid in comparison with real biological outcome prediction in the steady-state domain. Our technique produces biologically faithful models without the need for rate kinetics, detailed timing information, or complex inference procedures. To demonstrate the method, we produce a model using 28 pathways from the biological literature pertaining to the transcription factor family nuclear factor-κB. Predictions from this model in the steady-state domain are then validated against nine mice knockout experiments.

Inferring gene expression from ribosomal promoter sequences, a crowdsourcing approach

The Gene Promoter Expression Prediction challenge consisted of predicting gene expression from promoter sequences in a previously unknown experimentally generated data set. The challenge was presented to the community in the framework of the sixth Dialogue for Reverse Engineering Assessments and Methods (DREAM6), a community effort to evaluate the status of systems biology modeling methodologies. Nucleotide-specific promoter activity was obtained by measuring fluorescence from promoter sequences fused upstream of a gene for yellow fluorescence protein and inserted in the same genomic site of yeast Saccharomyces cerevisiae. Twenty-one teams submitted results predicting the expression levels of 53 different promoters from yeast ribosomal protein genes. Analysis of participant predictions shows that accurate values for low-expressed and mutated promoters were difficult to obtain, although in the latter case, only when the mutation induced a large change in promoter activity compared to the wild-type sequence. As in previous DREAM challenges, we found that aggregation of participant predictions provided robust results, but did not fare better than the three best algorithms. Finally, this study not only provides a benchmark for the assessment of methods predicting activity of a specific set of promoters from their sequence, but it also shows that the top performing algorithm, which used machine-learning approaches, can be improved by the addition of biological features such as transcription factor binding sites.

A simple approach to overcoming mutual coupling effects in some transmit array coils for magnetic resonance imaging

Single Echo Acquisition (SEA) is a method of completely parallel MR imaging that uses coil elements for spatial localization during receive, replacing the need for phase encoding repetitions. In this receive-only form, SEA imaging requires the use of a phase compensation gradient, the value of which is dependent on coil geometry, imaging distance from the elements, and element orientation. Operation of the arrays in transmit-receive mode, while adding significant complexity, is one potential method of eliminating the restrictions imposed by the phase compensation gradient. This abstract examines a straightforward current-splitting technique to enable parallel transmission for studying the complicated field interactions of these array coils in transmit mode.

Comprehensive site-specific whole genome profiling of stromal and epithelial colonic gene signatures in human sigmoid colon and rectal tissue

The strength of associations between various exposures (e.g., diet, tobacco, chemopreventive agents) and colorectal cancer risk may partially depend on the complex interaction between epithelium and stroma across anatomic subsites. Currently, baseline data describing genome-wide coding and long noncoding gene expression profiles in the healthy colon specific to tissue type and location are lacking. Therefore, colonic mucosal biopsies from 10 healthy participants who were enrolled in a clinical study to evaluate effects of lignan supplementation on gut resiliency were used to characterize the site-specific global gene expression signatures associated with stromal vs. epithelial cells in the sigmoid colon and rectum. Using RNA-seq, we demonstrate that tissue type and location patterns of gene expression and upstream regulatory pathways are distinct. For example, consistent with a key role of stroma in the crypt niche, mRNAs associated with immunoregulatory and inflammatory processes (i.e., CXCL14, ANTXR1), smooth muscle contraction (CALD1), proliferation and apoptosis (GLP2R, IGFBP3), and modulation of extracellular matrix (MMP2, COL3A1, MFAP4) were all highly expressed in the stroma. In comparison, HOX genes (HOXA3, HOXD9, HOXD10, HOXD11, and HOXD-AS2, a HOXD cluster antisense RNA 2), and WNT5B expression were also significantly higher in sigmoid colon compared with the rectum. These findings provide strong impetus for considering colorectal tissue subtypes and location in future observational studies and clinical trials designed to evaluate the effects of exposures on colonic health.

Attractor estimation and model refinement for stochastic regulatory network models

Using an expectation maximization procedure and reporter protein measurements from patients, we simultaneously refine a pre-existing stochastic regulatory network model and perform an attractor state estimation that will be used to improve therapeutic performance given the diseased tissue context of a particular patient.

Detecting Multivariate Gene Interactions in RNA-Seq Data Using Optimal Bayesian Classification

Differential gene expression testing is an analysis commonly applied to RNA-Seq data. These statistical tests identify genes that are significantly different across phenotypes. We extend this testing paradigm to multivariate gene interactions from a classification perspective with the goal to detect novel gene interactions for the phenotypes of interest. This is achieved through our novel computational framework comprised of a hierarchical statistical model of the RNA-Seq processing pipeline and the corresponding optimal Bayesian classifier. Through Markov Chain Monte Carlo sampling and Monte Carlo integration, we compute quantities where no analytical formulation exists. The performance is then illustrated on an expression dataset from a dietary intervention study where we identify gene pairs that have low classification error yet were not identified as differentially expressed. Additionally, we have released the software package to perform OBC classification on RNA-Seq data under an open source license and is available at http://bit.ly/obc_package.