Hierarchical Dirichlet process model for gene expression clustering

Population food intake clusters and cardiovascular disease incidence: a Bayesian quantifying of a prospective population-based cohort study in a low and middle-income country

Aims

This study was designed to explore the relationship between cardiovascular disease incidence and population clusters, which were established based on daily food intake.

Methods

The current study examined 5,396 Iranian adults (2,627 males and 2,769 females) aged 35 years and older, who participated in a 10-year longitudinal population-based study that began in 2001. The frequency of food group consumption over the preceding year (daily, weekly, or monthly) was assessed using a 49-item qualitative food frequency questionnaire (FFQ) administered via a face-to-face interview conducted by an expert dietitian. Participants were clustered based on their dietary intake by applying the semi-parametric Bayesian approach of the Dirichlet Process. In this approach, individuals with the same multivariate distribution based on dietary intake were assigned to the same cluster. The association between the extracted population clusters and the incidence of cardiovascular diseases was examined using Cox proportional hazard models.

Results

In the 10-year follow-up, 741 participants (401 men and 340 women) were diagnosed with cardiovascular diseases. Individuals were categorized into three primary dietary clusters: healthy, unhealthy, and mixed. After adjusting for potential confounders, subjects in the unhealthy cluster exhibited a higher risk for cardiovascular diseases [Hazard Ratio (HR): 2.059; 95% CI: 1.013, 4.184] compared to those in the healthy cluster. In the unadjusted model, individuals in the mixed cluster demonstrated a higher risk for cardiovascular disease than those in the healthy cluster (HR: 1.515; 95% CI: 1.097, 2.092). However, this association was attenuated after adjusting for potential confounders (HR: 1.145; 95% CI: 0.769, 1.706).

Conclusion

The results have shown that individuals within an unhealthy cluster have a risk that is twice as high for the incidence of cardiovascular diseases. However, these associations need to be confirmed through further prospective investigations.

Unsupervised Grouped Axial Data Modeling via Hierarchical Bayesian Nonparametric Models With Watson Distributions

This paper aims at proposing an unsupervised hierarchical nonparametric Bayesian framework for modeling axial data (i.e., observations are axes of direction) that can be partitioned into multiple groups, where each observation within a group is sampled from a mixture of Watson distributions with an infinite number of components that are allowed to be shared across different groups. First, we propose a hierarchical nonparametric Bayesian model for modeling grouped axial data based on the hierarchical Pitman-Yor process mixture model of Watson distributions. Then, we demonstrate that by setting the discount parameters of the proposed model to 0, another hierarchical nonparametric Bayesian model based on hierarchical Dirichlet process can be derived for modeling axial data. To learn the proposed models, we systematically develop a closed-form optimization algorithm based on the collapsed variational Bayes (CVB) inference. Furthermore, to ensure the convergence of the proposed learning algorithm, an annealing mechanism is introduced to the framework of CVB inference, leading to an averaged collapsed variational Bayes inference strategy. The merits of the proposed models for modeling grouped axial data are demonstrated through experiments on both synthetic data and real-world applications involving gene expression data clustering and depth image analysis.

Dirichlet process mixture models for single-cell RNA-seq clustering

Clustering of cells based on gene expression is one of the major steps in single-cell RNA-sequencing (scRNA-seq) data analysis. One key challenge in cluster analysis is the unknown number of clusters and, for this issue, there is still no comprehensive solution. To enhance the process of defining meaningful cluster resolution, we compare Bayesian latent Dirichlet allocation (LDA) method to its non-parametric counterpart, hierarchical Dirichlet process (HDP) in the context of clustering scRNA-seq data. A potential main advantage of HDP is that it does not require the number of clusters as an input parameter from the user. While LDA has been used in single-cell data analysis, it has not been compared in detail with HDP. Here, we compare the cell clustering performance of LDA and HDP using four scRNA-seq datasets (immune cells, kidney, pancreas and decidua/placenta), with a specific focus on cluster numbers. Using both intrinsic (DB-index) and extrinsic (ARI) cluster quality measures, we show that the performance of LDA and HDP is dataset dependent. We describe a case where HDP produced a more appropriate clustering compared to the best performer from a series of LDA clusterings with different numbers of clusters. However, we also observed cases where the best performing LDA cluster numbers appropriately capture the main biological features while HDP tended to inflate the number of clusters. Overall, our study highlights the importance of carefully assessing the number of clusters when analyzing scRNA-seq data.

Summary: Dirichlet mixture models (LDA and HDP) are applied for clustering cells in scRNA-Seq data. Here we made a comprehensive comparison of LDA and HDP model-based clustering for scRNA-seq data.

Variable Selection in Switching Dynamic Regression Models

Complex dynamic phenomena in which dynamics is related to events (modes) that cause structural changes over time, are well described by the switching linear dynamical system (SLDS). We extend the SLDS by allowing the measurement noise to be mode-specific, a flexible way to model non stationary data. Additionally, for models that are functions of explanatory variables, we adapt a variable selection method to identify which of them are significant in each mode. Our proposed model is a flexible Bayesian nonparametric model that allows to learn about the number of modes and their location, and within each mode, it identifies the significant variables and estimates the regression coefficients. The model performance is evaluated by simulation and two application examples from a dataset of meteorological time series of Barranquilla, Colombia are presented.

The Use of Informed Priors in Biclustering of Gene Expression with the Hierarchical Dirichlet Process

We motivate and describe the application of Hierarchical Dirichlet Process (HDP) models to the "soft" biclustering of gene expression data, in which we obtain modules (biclusters) where the affiliation of genes and samples with the modules are weighted, instead of being hard memberships. As a distinct contribution, we propose a method which HDP is informed with prior beliefs, significantly increasing the quality of the biclustering in terms of both the correctness of the number of modules inferred, and the precision of these modules, especially when evidence is sparse. We outline two such informed priors; one based on co-expression relationships inherent in the data, the other based on an externally provided regulatory network. We validate these results and compare the performance of our approach to Weighted Gene Correlation Network Analysis (WGCNA), another model that features weighted modules. We have, to this end, performed experiments on semi-synthetic data. The results show that HDP, with the addition of a well-informed prior, is able to capture the correct number of modules with increased accuracy. Furthermore, the model becomes robust to changes in the strength of the prior. We conclude by discussing these results and the benefits provided by our approach for gene expression analysis and network validation.

Compressing gene expression data using multiple latent space dimensionalities learns complementary biological representations

BACKGROUND

Unsupervised compression algorithms applied to gene expression data extract latent or hidden signals representing technical and biological sources of variation. However, these algorithms require a user to select a biologically appropriate latent space dimensionality. In practice, most researchers fit a single algorithm and latent dimensionality. We sought to determine the extent by which selecting only one fit limits the biological features captured in the latent representations and, consequently, limits what can be discovered with subsequent analyses.

RESULTS

We compress gene expression data from three large datasets consisting of adult normal tissue, adult cancer tissue, and pediatric cancer tissue. We train many different models across a large range of latent space dimensionalities and observe various performance differences. We identify more curated pathway gene sets significantly associated with individual dimensions in denoising autoencoder and variational autoencoder models trained using an intermediate number of latent dimensionalities. Combining compressed features across algorithms and dimensionalities captures the most pathway-associated representations. When trained with different latent dimensionalities, models learn strongly associated and generalizable biological representations including sex, neuroblastoma MYCN amplification, and cell types. Stronger signals, such as tumor type, are best captured in models trained at lower dimensionalities, while more subtle signals such as pathway activity are best identified in models trained with more latent dimensionalities.

CONCLUSIONS

There is no single best latent dimensionality or compression algorithm for analyzing gene expression data. Instead, using features derived from different compression models across multiple latent space dimensionalities enhances biological representations.

Biomarker Identification for Cancer Disease Using Biclustering Approach: An Empirical Study

This paper presents an exhaustive empirical study to identify biomarkers using two approaches: frequency-based and network-based, over seventeen different biclustering algorithms and six different cancer expression datasets. To systematically analyze the biclustering algorithms, we perform enrichment analysis, subtype identification and biomarker identification. Biclustering algorithms such as C&C, SAMBA and Plaid are useful to detect biomarkers by both approaches for all datasets except prostate cancer. We detect a total of 102 gene biomarkers using frequency-based method out of which 19 are for blood cancer, 36 for lung cancer, 25 for colon cancer, 13 for multi-tissue cancer and 9 for prostate cancer. Using the network-based approach we detect a total of 41 gene biomarkers of which 15 are from blood cancer, 12 from lung cancer, 6 from colon cancer, 7 from multi-tissue cancer and 1 from prostate cancer dataset. We further extend our network analysis over some biclusters and detect some gene biomarkers not detected earlier by both frequency-based or network-based approach. We expand our work on breast cancer miRNA expression data to evaluate the performance of the biclustering algorithms. We detect 19 breast cancer biomarkers by frequency-based method and 5 by network-based method for the miRNA dataset.

An application of Dirichlet process in clustering subjects via variance shift models: A course-evaluation study

In this article, the Dirichlet process (DP) is applied to cluster subjects with longitudinal observations. The basis of clustering is the ability of subjects to adapt themselves to new circumstances. Indeed, the basis of clustering depends on the time of changing response variability. This is done by providing a random change-point time in the variance structure of mixed-effects models. The DP is assumed as a prior for the distribution of the random change point. The discrete nature of the DP is utilized to cluster subjects according to the time of adaption. The proposed model is useful to identify groups of subjects with distinctive time-based progressions or declines. Transition mixed-effects models are also used to account for the serial correlation among observations over time. A joint modelling approach is utilized to handle the bias created in these models. The Gibbs sampling technique is adopted to achieve parameter estimates. Performance of the proposed method is evaluated via conducting a simulation study. The usefulness of the proposed model is assessed on a course-evaluation dataset.

Clustering Algorithms: Their Application to Gene Expression Data

Gene expression data hide vital information required to understand the biological process that takes place in a particular organism in relation to its environment. Deciphering the hidden patterns in gene expression data proffers a prodigious preference to strengthen the understanding of functional genomics. The complexity of biological networks and the volume of genes present increase the challenges of comprehending and interpretation of the resulting mass of data, which consists of millions of measurements; these data also inhibit vagueness, imprecision, and noise. Therefore, the use of clustering techniques is a first step toward addressing these challenges, which is essential in the data mining process to reveal natural structures and identify interesting patterns in the underlying data. The clustering of gene expression data has been proven to be useful in making known the natural structure inherent in gene expression data, understanding gene functions, cellular processes, and subtypes of cells, mining useful information from noisy data, and understanding gene regulation. The other benefit of clustering gene expression data is the identification of homology, which is very important in vaccine design. This review examines the various clustering algorithms applicable to the gene expression data in order to discover and provide useful knowledge of the appropriate clustering technique that will guarantee stability and high degree of accuracy in its analysis procedure.

DGEclust: differential expression analysis of clustered count data

We present a statistical methodology, DGEclust, for differential expression analysis of digital expression data. Our method treats differential expression as a form of clustering, thus unifying these two concepts. Furthermore, it simultaneously addresses the problem of how many clusters are supported by the data and uncertainty in parameter estimation. DGEclust successfully identifies differentially expressed genes under a number of different scenarios, maintaining a low error rate and an excellent control of its false discovery rate with reasonable computational requirements. It is formulated to perform particularly well on low-replicated data and be applicable to multi-group data. DGEclust is available at http://dvav.github.io/dgeclust/.