Search

Background: A fundamental problem in RNA-seq data analysis is to identify genes or exons that are differentially expressed with varying experimental conditions based on the read counts. The relativeness of RNA-seq measurements makes the between-sample normalization of read counts an essential step in differential expression (DE) analysis. In most existing methods, the normalization step is performed prior to the DE analysis. Recently, Jiang and Zhan proposed a statistical method which introduces sample-specific normalization parameters into a joint model, which allows for simultaneous normalization and differential expression analysis from log-transformed RNA-seq data. Furthermore, an ℓ 0 penalty is used to yield a sparse solution which selects a subset of DE genes. The experimental conditions are restricted to be categorical in their work., Results: In this paper, we generalize Jiang and Zhan's method to handle experimental conditions that are measured in continuous variables. As a result, genes with expression levels associated with a single or multiple covariates can be detected. As the problem being high-dimensional, non-differentiable and non-convex, we develop an efficient algorithm for model fitting., Conclusions: Experiments on synthetic data demonstrate that the proposed method outperforms existing methods in terms of detection accuracy when a large fraction of genes are differentially expressed in an asymmetric manner, and the performance gain becomes more substantial for larger sample sizes. We also apply our method to a real prostate cancer RNA-seq dataset to identify genes associated with pre-operative prostate-specific antigen (PSA) levels in patients.

Brain imaging genetics intends to uncover associations between genetic markers and neuroimaging quantitative traits. Sparse canonical correlation analysis (SCCA) can discover bi-multivariate associations and select relevant features, and is becoming popular in imaging genetic studies. The L1-norm function is not only convex, but also singular at the origin, which is a necessary condition for sparsity. Thus most SCCA methods impose [Formula: see text]-norm onto the individual feature or the structure level of features to pursuit corresponding sparsity. However, the [Formula: see text]-norm penalty over-penalizes large coefficients and may incurs estimation bias. A number of non-convex penalties are proposed to reduce the estimation bias in regression tasks. But using them in SCCA remains largely unexplored. In this paper, we design a unified non-convex SCCA model, based on seven non-convex functions, for unbiased estimation and stable feature selection simultaneously. We also propose an efficient optimization algorithm. The proposed method obtains both higher correlation coefficients and better canonical loading patterns. Specifically, these SCCA methods with non-convex penalties discover a strong association between the APOE e4 rs429358 SNP and the hippocampus region of the brain. They both are Alzheimer's disease related biomarkers, indicating the potential and power of the non-convex methods in brain imaging genetics.

Brain imaging genetics attracts more and more attention since it can reveal associations between genetic factors and the structures or functions of human brain. Sparse canonical correlation analysis (SCCA) is a powerful bi-multivariate association identification technique in imaging genetics. There have been many SCCA methods which could capture different types of structured imaging genetic relationships. These methods either use the group lasso to recover the group structure, or employ the graph/network guided fused lasso to find out the network structure. However, the group lasso methods have limitation in generalization because of the incomplete or unavailable prior knowledge in real world. The graph/network guided methods are sensitive to the sign of the sample correlation which may be incorrectly estimated. We introduce a new SCCA model using a novel graph guided pairwise group lasso penalty, and propose an efficient optimization algorithm. The proposed method has a strong upper bound for the grouping effect for both positively and negatively correlated variables. We show that our method performs better than or equally to two state-of-the-art SCCA methods on both synthetic and real neuroimaging genetics data. In particular, our method identifies stronger canonical correlations and captures better canonical loading profiles, showing its promise for revealing biologically meaningful imaging genetic associations.

Artificial neural networks (ANNs), like convolutional neural networks (CNNs), have achieved the state-of-the-art results for many machine learning tasks. However, inference with large-scale full-precision CNNs must cause substantial energy consumption and memory occupation, which seriously hinders their deployment on mobile and embedded systems. Highly inspired from biological brain, spiking neural networks (SNNs) are emerging as new solutions because of natural superiority in brain-like learning and great energy efficiency with event-driven communication and computation. Nevertheless, training a deep SNN remains a main challenge and there is usually a big accuracy gap between ANNs and SNNs. In this paper, we introduce a hardware-friendly conversion algorithm called "scatter-and-gather" to convert quantized ANNs to lossless SNNs, where neurons are connected with ternary {−1,0,1} synaptic weights. Each spiking neuron is stateless and more like original McCulloch and Pitts model, because it fires at most one spike and need be reset at each time step. Furthermore, we develop an incremental mapping framework to demonstrate efficient network deployments on a reconfigurable neuromorphic chip. Experimental results show our spiking LeNet on MNIST and VGG-Net on CIFAR-10 datasetobtain 99.37% and 91.91% classification accuracy, respectively. Besides, the presented mapping algorithm manages network deployment on our neuromorphic chip with maximum resource efficiency and excellent flexibility. Our four-spike LeNet and VGG-Net on chip can achieve respective real-time inference speed of 0.38 ms/image, 3.24 ms/image, and an average power consumption of 0.28 mJ/image and 2.3 mJ/image at 0.9 V, 252 MHz, which is nearly two orders of magnitude more efficient than traditional GPUs. [ABSTRACT FROM AUTHOR]

Subspace-based high-resolution algorithms for direction-of-arrival (DOA) estimation have been developed for large-scale adaptive antenna arrays. However, its prerequisite step, namely, source enumeration, has not yet been addressed. In this paper, a new approach is devised in the framework of the Bayesian information criterion (BIC) to provide reliable detection of the signal source number for the general asymptotic regime, where m,n\rightarrow\infty and m/n\rightarrow c\in(\mbox{0},\infty), with m$ and $n$ being the numbers of antennas and snapshots, respectively. In particular, the a posteriori probability is determined by correctly calculating the LLFs and PFs for the general asymptotic case. By means of the maximum a posteriori probability, we are capable of effectively finding the signal number. An accurate closed-form expression for the probability of missed detection is also derived for the proposed BIC variant. In addition, the probability of false alarm for the BIC detector is proved to converge to zero as $m,n\!\rightarrow\!\infty$ and $m/n\!\rightarrow\! c$. Simulation results are included to demonstrate the superiority of the proposed detection approach over state-of-the-art schemes and corroborate our theoretical calculations. [ABSTRACT FROM AUTHOR]