Most analyses of microarray data are based on point estimates of expression levels and ignore the uncertainty of such estimates. By determining uncertainties from Affymetrix GeneChip data and propagating these uncertainties to downstream analyses it has been shown that we can improve results of differential expression detection, principal component analysis and clustering. Previously, implementations of these uncertainty propagation methods have only been available as separate packages, written in different languages. Previous implementations have also suffered from being very costly to compute, and in the case of differential expression detection, have been limited in the experimental designs to which they can be applied.
Microarray technology is associated with many sources of experimental uncertainty. In this review we discuss a number of approaches for dealing with this uncertainty in the processing of data from microarray experiments. We focus here on the analysis of high-density oligonucleotide arrays, such as the popular Affymetrix GeneChip® array, which contain multiple probes for each target. This set of probes can be used to determine an estimate for the target concentration and can also be used to determine the experimental uncertainty associated with this measurement. This measurement uncertainty can then be propagated through the downstream analysis using probabilistic methods. We give examples showing how these credibility intervals can be used to help identify differential expression, to combine information from replicated experiments and to improve the performance of principal component analysis.
The high-throughput sequencing technology, RNA-Seq, has been widely used to quantify gene and isoform expression in the study of transcriptome in recent years. Accurate expression measurement from the millions or billions of short generated reads is obstructed by difficulties. One is ambiguous mapping of reads to reference transcriptome caused by alternative splicing. This increases the uncertainty in estimating isoform expression. The other is non-uniformity of read distribution along the reference transcriptome due to positional, sequencing, mappability and other undiscovered sources of biases. This violates the uniform assumption of read distribution for many expression calculation approaches, such as the direct RPKM calculation and Poisson-based models. Many methods have been proposed to address these difficulties. Some approaches employ latent variable models to discover the underlying pattern of read sequencing. However, most of these methods make bias correction based on surrounding sequence contents and share the bias models by all genes. They therefore cannot estimate gene- and isoform-specific biases as revealed by recent studies. We propose a latent variable model, NLDMseq, to estimate gene and isoform expression. Our method adopts latent variables to model the unknown isoforms, from which reads originate, and the underlying percentage of multiple spliced variants. The isoform- and exon-specific read sequencing biases are modeled to account for the non-uniformity of read distribution, and are identified by utilizing the replicate information of multiple lanes of a single library run. We employ simulation and real data to verify the performance of our method in terms of accuracy in the calculation of gene and isoform expression. Results show that NLDMseq obtains competitive gene and isoform expression compared to popular alternatives. Finally, the proposed method is applied to the detection of differential expression (DE) to show its usefulness in the downstream analysis. The proposed NLDMseq method provides an approach to accurately estimate gene and isoform expression from RNA-Seq data by modeling the isoform- and exon-specific read sequencing biases. It makes use of a latent variable model to discover the hidden pattern of read sequencing. We have shown that it works well in both simulations and real datasets, and has competitive performance compared to popular methods. The method has been implemented as a freely available software which can be found at https://github.com/PUGEA/NLDMseq .
Background: Clustering is an important analysis performed on microarray gene expression data since it groups genes which have similar expression patterns and enables the exploration of unknown gene functions. Microarray experiments are associated with many sources of experimental and biological variation and the resulting gene expression data are therefore very noisy. Many heuristic and model-based clustering approaches have been developed to cluster this noisy data. However, few of them include consideration of probe-level measurement error which provides rich information about technical variability.Results: We augment a standard model-based clustering method to incorporate probe-level measurement error. Using probe-level measurements from a recently developed Affymetrix probe-level model, multi-mgMOS, we include the probe-level measurement error directly into the standard Gaussian mixture model. Our augmented model is shown to provide improved clustering performance on simulated datasets and a real mouse time-course dataset.Conclusion: The performance of model-based clustering of gene expression data is improved by including probe-level measurement error and more biologically meaningful clustering results are obtained.
Mesh adaptation is crucial in numerical simulation, providing optimal resource allocation for accurately capturing physical phenomena. However, when applied to Computational Fluid Dynamics (CFD) problems with complex multi-scale properties, existing adaptation methods face huge challenges due to the high computational cost of solving auxiliary partial differential equations (PDEs) and the difficulty in aligning the flow features with mesh geometric features. In this work, an end-to-end data-driven mesh adaptation framework, Flow2Mesh, is proposed to address these challenges by adopting a hybrid modeling strategy to construct the mapping from pixelated flow-fields to graph-based meshes. It achieves a rapid and accurate one-step mesh adaptation via a perceptual feature network (PFN) and a mesh movement network (MMN). PFN extracts the global perceptual features from flow-fields to enhance flow feature representation and mesh resolution independence. In MMN, these features are utilized to deform the initial mesh to a topology-invariant adaptive mesh by a proposed physically driven mesh convolutional network. It considers the inherent mesh geometric information for efficient node feature aggregation and alignment of mesh density with a flow-field structure. To generate high-quality adaptive meshes, various mesh-related losses are designed to regularize the mesh movement and alleviate the mesh tangling. Experiments in CFD scenarios demonstrate the generalization of our model to different design parameters and mesh configurations. It takes three orders of magnitude less time to generate similar meshes than the PDE-based method. The results exhibit the potential of Flow2Mesh to be a flexible and reliable tool for rapid mesh adaptation in scientific and industrial fields.
Separation-induced transitions can affect the hydrodynamic performance of submarines significantly. The onset of this transition plays a critical role in the design of aircrafts and underwater vehicles. Since the transition is affected by various factors, its prediction is a challenging task. Convolutional neural networks (CNNs) in the field of machine learning can extract features from high-dimensional data automatically and have good generalization performance. In this paper, we propose an end-to-end CNN-based model to automatically extract the features of separation-induced transitions by learning from image-expressed flow field data. The proposed data-driven deep learning model employs a high-resolution network, which is widely used in key-point detection in the field of computer vision, to extract the underlying features in the separation-induced transition under only a few empirical assumptions. A novel representation of separation-induced transition onset in the form of a heatmap is especially proposed to indicate the probability of transition onset. We use implicit-large-eddy-simulation data generated by the second-order discontinuous Galerkin method over Lyu's-1 model line submarine to verify the proposed method, and the results demonstrate that the new method is able to predict separation-induced transition onsets with quantified uncertainty. The proposed model can be used as an auxiliary tool for aerodynamic and hydrodynamic designs.