Sanvesh Srivastava

Differential expression—the next generation and beyond

RNA-sequencing (RNA-seq) technologies have not only pushed the boundaries of science, but also pushed the computational and analytic capacities of many laboratories. With respect to mapping and quantifying transcriptomes, RNA-seq has certainly established itself as the approach of choice. However, as the complexities of experiments continue to grow, there is still no standard practice that allows for design, processing, normalization, efficient dimension reduction and/or statistical analysis. With this in mind, we provide a brief review of some of the key challenges that are general to all RNA-seq experiments, namely experimental design, statistical analysis and dimensionality reduction.

Essential role of the plasmid hik31 operon in regulating central metabolism in the dark in Synechocystis sp. PCC 6803.

The plasmid hik31 operon (P3, slr6039‐slr6041) is located on the pSYSX plasmid in Synechocystis sp. PCC 6803. A P3 mutant (ΔP3) had a growth defect in the dark and a pigment defect that was worsened by the addition of glucose. The glucose defect was from incomplete metabolism of the substrate, was pH dependent, and completely overcome by the addition of bicarbonate. Addition of organic carbon and nitrogen sources partly alleviated the defects of the mutant in the dark. Electron micrographs of the mutant revealed larger cells with division defects, glycogen limitation, lack of carboxysomes, deteriorated thylakoids and accumulation of polyhydroxybutyrate and cyanophycin. A microarray experiment over two days of growth in light‐dark plus glucose revealed downregulation of several photosynthesis, amino acid biosynthesis, energy metabolism genes; and an upregulation of cell envelope and transport and binding genes in the mutant. ΔP3 had an imbalance in carbon and nitrogen levels and many sugar catabolic and cell division genes were negatively affected after the first dark period. The mutant suffered from oxidative and osmotic stress, macronutrient limitation, and an energy deficit. Therefore, the P3 operon is an important regulator of central metabolism and cell division in the dark.

Simple, scalable and accurate posterior interval estimation

Summary Standard posterior sampling algorithms, such as Markov chain Monte Carlo procedures, face major challenges in scaling up to massive datasets. We propose a simple and general posterior interval estimation algorithm to rapidly and accurately estimate quantiles of the posterior distributions for one‐dimensional functionals. Our algorithm runs Markov chain Monte Carlo in parallel for subsets of the data, and then averages quantiles estimated from each subset. We provide strong theoretical guarantees and show that the credible intervals from our algorithm asymptotically approximate those from the full posterior in the leading parametric order. Our algorithm has a better balance of accuracy and efficiency than its competitors across a variety of simulations and a real‐data example.

Differential expression of candidate salivary effector proteins in field collections of Hessian fly, Mayetiola destructor

Evidence is emerging that some proteins secreted by gall‐forming parasites of plants act as effectors responsible for systemic changes in the host plant, such as galling and nutrient tissue formation. A large number of secreted salivary gland proteins (SSGPs) that are the putative effectors responsible for the physiological changes elicited in susceptible seedling wheat by Hessian fly, Mayetiola destructor (Say), larvae have been documented. However, how the genes encoding these candidate effectors might respond under field conditions is unknown. The goal of this study was to use microarray analysis to investigate variation in SSGP transcript abundance amongst field collections from different geographical regions (southeastern USA, central USA, and the Middle East). Results revealed significant variation in SSGP transcript abundance amongst the field collections studied. The field collections separated into three distinct groups that corresponded to the wheat classes grown in the different geographical regions as well as to recently described Hessian fly populations. These data support previous reports correlating Hessian fly population structure with micropopulation differences owing to agro‐ecosystem parameters such as cultivation of regionally adapted wheat varieties, deployment of resistance genes and variation in climatic conditions.

Rat intersubjective decisions are encoded by frequency-specific oscillatory contexts.

It is unknown how the brain coordinates decisions to withstand personal costs in order to prevent other individuals’ distress. Here we test whether local field potential (LFP) oscillations between brain regions create “neural contexts” that select specific brain functions and encode the outcomes of these types of intersubjective decisions.

Expandable factor analysis

Summary Bayesian sparse factor models have proven useful for characterizing dependence in multivariate data, but scaling computation to large numbers of samples and dimensions is problematic. We propose expandable factor analysis for scalable inference in factor models when the number of factors is unknown. The method relies on a continuous shrinkage prior for efficient maximum a posteriori estimation of a low-rank and sparse loadings matrix. The structure of the prior leads to an estimation algorithm that accommodates uncertainty in the number of factors. We propose an information criterion to select the hyperparameters of the prior. Expandable factor analysis has better false discovery rates and true positive rates than its competitors across diverse simulation settings. We apply the proposed approach to a gene expression study of ageing in mice, demonstrating superior results relative to four competing methods.

An Asynchronous Distributed Expectation Maximization Algorithm For Massive Data: The DEM Algorithm

ABSTRACT The family of expectation--maximization (EM) algorithms provides a general approach to fitting flexible models for large and complex data. The expectation (E) step of EM-type algorithms is time-consuming in massive data applications because it requires multiple passes through the full data. We address this problem by proposing an asynchronous and distributed generalization of the EM called the distributed EM (DEM). Using DEM, existing EM-type algorithms are easily extended to massive data settings by exploiting the divide-and-conquer technique and widely available computing power, such as grid computing. The DEM algorithm reserves two groups of computing processes called workers and managers for performing the E step and the maximization step (M step), respectively. The samples are randomly partitioned into a large number of disjoint subsets and are stored on the worker processes. The E step of DEM algorithm is performed in parallel on all the workers, and every worker communicates its results to the managers at the end of local E step. The managers perform the M step after they have received results from a γ-fraction of the workers, where γ is a fixed constant in (0, 1]. The sequence of parameter estimates generated by the DEM algorithm retains the attractive properties of EM: convergence of the sequence of parameter estimates to a local mode and linear global rate of convergence. Across diverse simulations focused on linear mixed-effects models, the DEM algorithm is significantly faster than competing EM-type algorithms while having a similar accuracy. The DEM algorithm maintains its superior empirical performance on a movie ratings database consisting of 10 million ratings. Supplementary material for this article is available online.

Integrating multi-platform genomic data using hierarchical Bayesian relevance vector machines

We present a statistical framework, hierarchical relevance vector machine (H-RVM), for improved prediction of scalar outcomes using interacting high-dimensional input covariates from different sources. We illustrate our methodology for integrating genomic data from multiple platforms to predict observed clinical phenotypes. H-RVM is a hierarchical Bayesian generalization of the relevance vector machine and its learning algorithm is a special case of the computationally efficient variational method of hierarchic kernel learning frame-work. We apply H-RVM to data from the Cancer Genome Atlas based Glioblastoma study to predict imaging-based tumor volume by integrating gene and miRNA expression data and show that H-RVM performs much better in prediction as compared to competing methods.

Purdue Statistics: A Journey Through Time

Since its modest inception as the Statistical Laboratory in 1947, the Department of Statistics, Purdue University has grown to one of the largest and most diverse in the country supported by a distinguished list of alumni, outstanding contributions in research, and major advances in statistical education. Its current (2011) size of 62 faculty, 125 graduate students, and almost 400 undergraduate students reflects its commitment to developing statistical sciences research for the present and the future, and to providing high quality education to students, both in statistics and in other disciplines. Historically, the Department of Statistics at Purdue University has been an important center for diverse areas of statistical research. Its strong presence in probability, theory, and education set the stage for its expansion in the mid-1990s. As the field of statistics expanded to include many interdisciplinary areas that require specialization (statistical genetics and bioinformatics, computational finance, machine learning, etc.), Purdue Statistics engaged in an aggressive program of hiring well-prepared faculty with diverse backgrounds who are playing leading roles in the development of the field as it expands its scope. Today Purdue Statistics stands strong as the highest ranked department in the College of Science at Purdue University, and is enjoying its place among the top ranked departments in the United States.