Brian E. Howard

High-Throughput RNA Sequencing of Pseudomonas-Infected Arabidopsis Reveals Hidden Transcriptome Complexity and Novel Splice Variants

We report the results of a genome-wide analysis of transcription in Arabidopsis thaliana after treatment with Pseudomonas syringae pathovar tomato. Our time course RNA-Seq experiment uses over 500 million read pairs to provide a detailed characterization of the response to infection in both susceptible and resistant hosts. The set of observed differentially expressed genes is consistent with previous studies, confirming and extending existing findings about genes likely to play an important role in the defense response to Pseudomonas syringae. The high coverage of the Arabidopsis transcriptome resulted in the discovery of a surprisingly large number of alternative splicing (AS) events – more than 44% of multi-exon genes showed evidence for novel AS in at least one of the probed conditions. This demonstrates that the Arabidopsis transcriptome annotation is still highly incomplete, and that AS events are more abundant than expected. To further refine our predictions, we identified genes with statistically significant changes in the ratios of alternative isoforms between treatments. This set includes several genes previously known to be alternatively spliced or expressed during the defense response, and it may serve as a pool of candidate genes for regulated alternative splicing with possible biological relevance for the defense response against invasive pathogens.

Towards reliable isoform quantification using RNA-SEQ data

BackgroundIn eukaryotes, alternative splicing often generates multiple splice variants from a single gene. Here weexplore the use of RNA sequencing (RNA-Seq) datasets to address the isoform quantification problem. Given a set of known splice variants, the goal is to estimate the relative abundance of the individual variants.MethodsOur method employs a linear models framework to estimate the ratios of known isoforms in a sample. A key feature of our method is that it takes into account the non-uniformity of RNA-Seq read positions along the targeted transcripts.ResultsPreliminary tests indicate that the model performs well on both simulated and real data. In two publicly available RNA-Seq datasets, we identified several alternatively-spliced genes with switch-like, on/off expression properties, as well as a number of other genes that varied more subtly in isoform expression. In many cases, genes exhibiting differential expression of alternatively spliced transcripts were not differentially expressed at the gene level.ConclusionsGiven that changes in isoform expression level frequently involve a continuum of isoform ratios, rather than all-or-nothing expression, and that they are often independent of general gene expression changes, we anticipate that our research will contribute to revealing a so far uninvestigated layer of the transcriptome. We believe that, in the future, researchers will prioritize genes for functional analysis based not only on observed changes in gene expression levels, but also on changes in alternative splicing.

Interaction of Temperature and Photoperiod Increases Growth and Oil Content in the Marine Microalgae Dunaliella viridis

Eukaryotic marine microalgae like Dunaliella spp. have great potential as a feedstock for liquid transportation fuels because they grow fast and can accumulate high levels of triacylgycerides with little need for fresh water or land. Their growth rates vary between species and are dependent on environmental conditions. The cell cycle, starch and triacylglycerol accumulation are controlled by the diurnal light:dark cycle. Storage compounds like starch and triacylglycerol accumulate in the light when CO2 fixation rates exceed the need of assimilated carbon and energy for cell maintenance and division during the dark phase. To delineate environmental effects, we analyzed cell division rates, metabolism and transcriptional regulation in Dunaliella viridis in response to changes in light duration and growth temperatures. Its rate of cell division was increased under continuous light conditions, while a shift in temperature from 25°C to 35°C did not significantly affect the cell division rate, but increased the triacylglycerol content per cell several-fold under continuous light. The amount of saturated fatty acids in triacylglycerol fraction was more responsive to an increase in temperature than to a change in the light regime. Detailed fatty acid profiles showed that Dunaliella viridis incorporated lauric acid (C12:0) into triacylglycerol after 24 hours under continuous light. Transcriptome analysis identified potential regulators involved in the light and temperature-induced lipid accumulation in Dunaliella viridis.

Mining of cis-Regulatory Motifs Associated with Tissue-Specific Alternative Splicing

Alternative splicing (AS) is an important post-transcriptional mechanism that can increase protein diversity and affect mRNA stability and translation efficiency. Many studies targeting the regulation of alternative splicing have focused on individual motifs; however, little is known about how such motifs work in concert. In this paper, we use distribution-based quantitative association rule mining to find combinatorial cis -regulatory motifs and to investigate the effect of motif pairs. We also show that motifs that occur in motif pairs typically occur in clusters.

A hybrid gene selection approach to create the S1500+ targeted gene sets for use in high-throughput transcriptomics

Changes in gene expression can help reveal the mechanisms of disease processes and the mode of action for toxicities and adverse effects on cellular responses induced by exposures to chemicals, drugs and environment agents. The U.S. Tox21 Federal collaboration, which currently quantifies the biological effects of nearly 10,000 chemicals via quantitative high-throughput screening(qHTS) in in vitro model systems, is now making an effort to incorporate gene expression profiling into the existing battery of assays. Whole transcriptome analyses performed on large numbers of samples using microarrays or RNA-Seq is currently cost-prohibitive. Accordingly, the Tox21 Program is pursuing a high-throughput transcriptomics (HTT) method that focuses on the targeted detection of gene expression for a carefully selected subset of the transcriptome that potentially can reduce the cost by a factor of 10-fold, allowing for the analysis of larger numbers of samples. To identify the optimal transcriptome subset, genes were sought that are (1) representative of the highly diverse biological space, (2) capable of serving as a proxy for expression changes in unmeasured genes, and (3) sufficient to provide coverage of well described biological pathways. A hybrid method for gene selection is presented herein that combines data-driven and knowledge-driven concepts into one cohesive method. Our approach is modular, applicable to any species, and facilitates a robust, quantitative evaluation of performance. In particular, we were able to perform gene selection such that the resulting set of “sentinel genes” adequately represents all known canonical pathways from Molecular Signature Database (MSigDB v4.0) and can be used to infer expression changes for the remainder of the transcriptome. The resulting computational model allowed us to choose a purely data-driven subset of 1500 sentinel genes, referred to as the S1500 set, which was then augmented using a knowledge-driven selection of additional genes to create the final S1500+ gene set. Our results indicate that the sentinel genes selected can be used to accurately predict pathway perturbations and biological relationships for samples under study.

Towards Reliable Isoform Quantification Using RNA-Seq Data

In eukaryotes, alternative splicing often generates multiple splice variants from a single gene. Here we explore the use of RNA sequencing (RNA-Seq) datasets to address the isoform quantification problem. Given a set of known splice variants, the goal is to estimate the relative abundance of the individual variants. Our method employs a linear models framework to estimate the ratios of known isoforms in a sample. A key feature of our method is that it takes into account the non-uniformity of RNA-Seq read positions along the targeted transcripts. Preliminary tests indicate that the model performs well on both simulated and real data.

Quality Assessment of Affymetrix GeneChip Data using the EM Algorithm and a Naive Bayes Classifier

Recent research has demonstrated the utility of using supervised classification systems for automatic identification of low quality microarray data. However, this approach requires annotation of a large training set by a qualified expert. In this paper we demonstrate the utility of an unsupervised classification technique based on the Expectation-Maximization (EM) algorithm and naive Bayes classification. On our test set, this system exhibits performance comparable to that of an analogous supervised learner constructed from the same training data.

Workshop: Using a transcript catalog and paired-end RNA-Seq data to identify differential alternative splicing

Alternative splicing is one of the major contributors to transcript and protein diversity in many higher eukaryotes including humans, but so far the extent and impact of alternative splicing in plants has not been thoroughly investigated [1]. The purpose of our NSF-Eager grant is to use high-throughput RNA sequencing of the transcriptome to assess the role of alternative splicing in the host-pathogen interaction between Arabidopsis thaliana and Pseudomonas syringae. Although previous studies have demonstrated that alternative splicing plays an important role for individual genes during this interaction [2, 3], our whole-genome approach will investigate thousands of potential splicing events simultaneously. Our method employs a linear models framework to estimate the ratios of known isoforms in a given sample, taking into account the non-uniformity of RNA-Sequencing reads along the targeted transcripts [4]. Recently, we have adapted our method in order to accommodate paired-end sequencing technology. Briefly, we map reads to a transcript catalog (TAIR 10 gene models), partition the reads according to their compatibility with these models, identify reads that are informative for isoform quantification, and then use a linear model to estimate isoform expression ratios for each gene. Our results provide a first approximation of the extent of pathogen-induced alternative splicing and so far indicate evidence for a wide variety of novel alternative transcripts. In addition, the set of differentially spliced genes appears to be independent of the set of differentially expressed genes, providing new evidence for a hidden layer of regulation in the transcriptome [5]. Hence, the set of differentially spliced genes provides a promising set of candidate genes that researchers may have previously overlooked by focusing solely on differential gene expression.

Improved RNA-Seq Partitions in Linear Models for Isoform Quantification

Here, we present an extension of our is form quantification method that accommodates paired end RNA Sequencing data. We explore several alternate methods of partitioning read count data in order to better exploit the available fragment size distribution, and to reduce the variance in the resulting estimates. In many cases, this significantly improves the accuracy of our approach.

Practical Quality Assessment of Microarray Data by Simulation of Differential Gene Expression

There are many methods for assessing the quality of microarray data, but little guidance regarding what to do when defective data is identified. Depending on the scientific question asked, discarding flawed data from a small experiment may be detrimental. Here we describe a novel quality assessment method that is designed to identify chips that should be discarded from an experiment. This technique simulates a set of differentially expressed genes and then assesses whether discarding each chip enhances or obscures the recovery of this known set. We compare our method to expert annotations derived using popular quality diagnostics and show, with examples, that the decision to discard a chip depends on the details of the particular experiment.