Neuron enriched extracellular vesicles' MicroRNA expression profiles as a marker of early life alcohol consumption.

Alcohol consumption may impact and shape brain development through perturbed biological pathways and impaired molecular functions. We investigated the relationship between alcohol consumption rates and neuron-enriched extracellular vesicles' (EVs') microRNA (miRNA) expression to better understand the impact of alcohol use on early life brain biology. Neuron-enriched EVs' miRNA expression was measured from plasma samples collected from young people using a commercially available microarray platform while alcohol consumption was measured using the Alcohol Use Disorders Identification Test. Linear regression and network analyses were used to identify significantly differentially expressed miRNAs and to characterize the implicated biological pathways, respectively. Compared to alcohol naïve controls, young people reporting high alcohol consumption exhibited significantly higher expression of three neuron-enriched EVs' miRNAs including miR-30a-5p, miR-194-5p, and miR-339-3p, although only miR-30a-5p and miR-194-5p survived multiple test correction. The miRNA-miRNA interaction network inferred by a network inference algorithm did not detect any differentially expressed miRNAs with a high cutoff on edge scores. However, when the cutoff of the algorithm was reduced, five miRNAs were identified as interacting with miR-194-5p and miR-30a-5p. These seven miRNAs were associated with 25 biological functions; miR-194-5p was the most highly connected node and was highly correlated with the other miRNAs in this cluster. Our observed association between neuron-enriched EVs' miRNAs and alcohol consumption concurs with results from experimental animal models of alcohol use and suggests that high rates of alcohol consumption during the adolescent/young adult years may impact brain functioning and development by modulating miRNA expression.

Neuron Enriched Exosomal MicroRNA Expression Profiles as a Marker of Early Life Alcohol Consumption.

Background: Alcohol consumption may impact and shape brain development through perturbed biological pathways and impaired molecular functions. We investigated the relationship between alcohol consumption rates and neuron-enriched exosomal microRNA (miRNA) expression to better understand the impact of alcohol use on early life brain biology. Methods: Neuron-enriched exosomal miRNA expression was measured from plasma samples collected from young people using a commercially available microarray platform while alcohol consumption was measured using the Alcohol Use Disorders Identification Test. Linear regression and network analyses were used to identify significantly differentially expressed miRNAs and to characterize the implicated biological pathways, respectively. Results: Compared to alcohol naïve controls, young people reporting high alcohol consumption exhibited significantly higher expression of four neuron-enriched exosomal miRNAs including miR-30a-5p, miR-194-5p, and miR-339-3p, although only miR-30a-5p and miR-194-5p survived multiple test correction. The miRNA-miRNA interaction network inferred by a network inference algorithm did not detect any differentially expressed miRNAs with a high cutoff on edge scores. However, when the cutoff of the algorithm was reduced, five miRNAs were identified as interacting with miR-194-5p and miR-30a-5p. These seven miRNAs were associated with 25 biological functions; miR-194-5p was the most highly connected node and was highly correlated with the other miRNAs in this cluster. Conclusions: Our observed association between neuron-enriched exosomal miRNAs and alcohol consumption concurs with results from experimental animal models of alcohol use and suggests that high rates of alcohol consumption during the adolescent/young adult years may impact brain functioning and development by modulating miRNA expression.

Relevant and Non-Redundant Feature Selection for Cancer Classification and Subtype Detection.

Biologists seek to identify a small number of significant features that are important, non-redundant, and relevant from diverse omics data. For example, statistical methods such as LIMMA and DEseq distinguish differentially expressed genes between a case and control group from the transcript profile. Researchers also apply various column subset selection algorithms on genomics datasets for a similar purpose. Unfortunately, genes selected by such statistical or machine learning methods are often highly co-regulated, making their performance inconsistent. Here, we introduce a novel feature selection algorithm that selects highly disease-related and non-redundant features from a diverse set of omics datasets. We successfully applied this algorithm to three different biological problems: (a) disease-to-normal sample classification; (b) multiclass classification of different disease samples; and (c) disease subtypes detection. Considering the classification of ROC-AUC, false-positive, and false-negative rates, our algorithm outperformed other gene selection and differential expression (DE) methods for all six types of cancer datasets from TCGA considered here for binary and multiclass classification problems. Moreover, genes picked by our algorithm improved the disease subtyping accuracy for four different cancer types over state-of-the-art methods. Hence, we posit that our proposed feature reduction method can support the community to solve various problems, including the selection of disease-specific biomarkers, precision medicine design, and disease sub-type detection.

Performance Comparison of Deep Learning Autoencoders for Cancer Subtype Detection Using Multi-Omics Data.

A heterogeneous disease such as cancer is activated through multiple pathways and different perturbations. Depending upon the activated pathway(s), the survival of the patients varies significantly and shows different efficacy to various drugs. Therefore, cancer subtype detection using genomics level data is a significant research problem. Subtype detection is often a complex problem, and in most cases, needs multi-omics data fusion to achieve accurate subtyping. Different data fusion and subtyping approaches have been proposed over the years, such as kernel-based fusion, matrix factorization, and deep learning autoencoders. In this paper, we compared the performance of different deep learning autoencoders for cancer subtype detection. We performed cancer subtype detection on four different cancer types from The Cancer Genome Atlas (TCGA) datasets using four autoencoder implementations. We also predicted the optimal number of subtypes in a cancer type using the silhouette score and found that the detected subtypes exhibit significant differences in survival profiles. Furthermore, we compared the effect of feature selection and similarity measures for subtype detection. For further evaluation, we used the Glioblastoma multiforme (GBM) dataset and identified the differentially expressed genes in each of the subtypes. The results obtained are consistent with other genomic studies and can be corroborated with the involved pathways and biological functions. Thus, it shows that the results from the autoencoders, obtained through the interaction of different datatypes of cancer, can be used for the prediction and characterization of patient subgroups and survival profiles.

Co-Expression Networks for Causal Gene Identification Based on RNA-Seq Data of Corynebacterium pseudotuberculosis.

Corynebacterium pseudotuberculosis is a Gram-positive bacterium that causes caseous lymphadenitis, a disease that predominantly affects sheep, goat, cattle, buffalo, and horses, but has also been recognized in other animals. This bacterium generates a severe economic impact on countries producing meat. Gene expression studies using RNA-Seq are one of the most commonly used techniques to perform transcriptional experiments. Computational analysis of such data through reverse-engineering algorithms leads to a better understanding of the genome-wide complexity of gene interactomes, enabling the identification of genes having the most significant functions inferred by the activated stress response pathways. In this study, we identified the influential or causal genes from four RNA-Seq datasets from different stress conditions (high iron, low iron, acid, osmosis, and PH) in C. pseudotuberculosis, using a consensus-based network inference algorithm called miRsigand next identified the causal genes in the network using the miRinfluence tool, which is based on the influence diffusion model. We found that over 50% of the genes identified as influential had some essential cellular functions in the genomes. In the strains analyzed, most of the causal genes had crucial roles or participated in processes associated with the response to extracellular stresses, pathogenicity, membrane components, and essential genes. This research brings new insight into the understanding of virulence and infection by C. pseudotuberculosis.

A game-theoretic approach to deciphering the dynamics of amyloid-ß aggregation along competing pathways.

Aggregation of amyloid-ß (Aß) peptides is a significant event that underpins Alzheimer's disease (AD). Aß aggregates, especially the low-molecular weight oligomers, are the primary toxic agents in AD pathogenesis. Therefore, there is increasing interest in understanding their formation and behaviour. In this paper, we use our previously established results on heterotypic interactions between Aß and fatty acids (FAs) to investigate off-pathway aggregation under the control of FA concentrations to develop a mathematical framework that captures the mechanism. Our framework to define and simulate the competing on- and off-pathways of Aß aggregation is based on the principles of game theory. Together with detailed simulations and biophysical experiments, our models describe the dynamics involved in the mechanisms of Aß aggregation in the presence of FAs to adopt multiple pathways. Specifically, our reduced-order computations indicate that the emergence of off- or on-pathway aggregates are tightly controlled by a narrow set of rate constants, and one could alter such parameters to populate a particular oligomeric species. These models agree with the detailed simulations and experimental data on using FA as a heterotypic partner to modulate the temporal parameters. Predicting spatio-temporal landscape along competing pathways for a given heterotypic partner such as lipids is a first step towards simulating scenarios in which the generation of specific 'conformer strains' of Aß could be predicted. This approach could be significant in deciphering the mechanisms of amyloid aggregation and strain generation, which are ubiquitously observed in many neurodegenerative diseases.

Recent advances on constraint-based models by integrating machine learning.

Research that meaningfully integrates constraint-based modeling with machine learning is at its infancy but holds much promise. Here, we consider where machine learning has been implemented within the constraint-based modeling reconstruction framework and highlight the need to develop approaches that can identify meaningful features from large-scale data and connect them to biological mechanisms to establish causality to connect genotype to phenotype. We motivate the construction of iterative integrative schemes where machine learning can fine-tune the input constraints in a constraint-based model or contrarily, constraint-based model simulation results are analyzed by machine learning and reconciled with experimental data. This can iteratively refine a constraint-based model until there is consistency between experimental data, machine learning results, and constraint-based model simulations.

Evaluation of the Common Molecular Basis in Alzheimer's and Parkinson's Diseases.

Alzheimer's disease (AD) and Parkinson's disease (PD) are the most common neurodegenerative disorders related to aging. Though several risk factors are shared between these two diseases, the exact relationship between them is still unknown. In this paper, we analyzed how these two diseases relate to each other from the genomic, epigenomic, and transcriptomic viewpoints. Using an extensive literature mining, we first accumulated the list of genes from major genome-wide association (GWAS) studies. Based on these GWAS studies, we observed that only one gene (HLA-DRB5) was shared between AD and PD. A subsequent literature search identified a few other genes involved in these two diseases, among which SIRT1 seemed to be the most prominent one. While we listed all the miRNAs that have been previously reported for AD and PD separately, we found only 15 different miRNAs that were reported in both diseases. In order to get better insights, we predicted the gene co-expression network for both AD and PD using network analysis algorithms applied to two GEO datasets. The network analysis revealed six clusters of genes related to AD and four clusters of genes related to PD; however, there was very low functional similarity between these clusters, pointing to insignificant similarity between AD and PD even at the level of affected biological processes. Finally, we postulated the putative epigenetic regulator modules that are common to AD and PD.

Cause and consequence of Aß - Lipid interactions in Alzheimer disease pathogenesis.

Self-templating propagation of protein aggregate conformations is increasingly becoming a significant factor in many neurological diseases. In Alzheimer disease (AD), intrinsically disordered amyloid-ß (Aß) peptides undergo aggregation that is sensitive to environmental conditions. High-molecular weight aggregates of Aß that form insoluble fibrils are deposited as senile plaques in AD brains. However, low-molecular weight aggregates called soluble oligomers are known to be the primary toxic agents responsible for neuronal dysfunction. The aggregation process is highly stochastic involving both homotypic (Aß-Aß) and heterotypic (Aß with interacting partners) interactions. Two of the important members of interacting partners are membrane lipids and surfactants, to which Aß shows a perpetual association. Aß-membrane interactions have been widely investigated for more than two decades, and this research has provided a wealth of information. Although this has greatly enriched our understanding, the objective of this review is to consolidate the information from the literature that collectively showcases the unique phenomenon of lipid-mediated Aß oligomer generation, which has largely remained inconspicuous. This is especially important because Aß aggregate "strains" are increasingly becoming relevant in light of the correlations between the structure of aggregates and AD phenotypes. Here, we will focus on aspects of Aß-lipid interactions specifically from the context of how lipid modulation generates a wide variety of biophysically and biochemically distinct oligomer sub-types. This, we believe, will refocus our thinking on the influence of lipids and open new approaches in delineating the mechanisms of AD pathogenesis. This article is part of a Special Issue entitled: Protein Aggregation and Misfolding at the Cell Membrane Interface edited by Ayyalusamy Ramamoorthy.

Propagation of an Aß Dodecamer Strain Involves a Three-Step Mechanism and a Key Intermediate.

Proteinaceous deposits composed of fibrillar amyloid-ß (Aß) are the primary neuropathological hallmarks in Alzheimer disease (AD) brains. The nucleation-dependent aggregation of Aß is a stochastic process with frequently observed heterogeneity in aggregate size, structure, and conformation that manifests in fibril polymorphism. Emerging evidence indicates that polymorphic variations in Aß fibrils contribute to phenotypic diversity and the rate of disease progression in AD. We recently demonstrated that a dodecamer strain derived from synthetic Aß42 propagates to morphologically distinct fibrils and selectively induces cerebral amyloid angiopathy phenotype in transgenic mice. This report supports the growing contention that stable oligomer strains can influence phenotypic outcomes by faithful propagation of their structures. Although we determined the mechanism of dodecamer propagation on a mesoscopic scale, the molecular details of the microscopic reactions remained unknown. Here, we have dissected and evaluated individually the kinetics of macroscopic phases in aggregation to gain insight into the process of strain propagation. The bulk rates determined experimentally in each phase were used to build an ensemble kinetic simulation model, which confirmed our observation that dodecamer seeds initially grow by monomer addition toward the formation of a key intermediate. This is followed by conversion of the intermediate to fibrils by oligomer elongation and association mechanisms. Overall, this report reveals important insights into the molecular details of oligomer strain propagation involved in AD pathology.

Fatty Acid Concentration and Phase Transitions Modulate Aß Aggregation Pathways.

Aggregation of amyloid ß (Aß) peptides is a significant event that underpins Alzheimer disease (AD) pathology. Aß aggregates, especially the low-molecular weight oligomers, are the primary toxic agents in AD and hence, there is increasing interest in understanding their formation and behavior. Aggregation is a nucleation-dependent process in which the pre-nucleation events are dominated by Aß homotypic interactions. Dynamic flux and stochasticity during pre-nucleation renders the reactions susceptible to perturbations by other molecules. In this context, we investigate the heterotypic interactions between Aß and fatty acids (FAs) by two independent tool-sets such as reduced order modelling (ROM) and ensemble kinetic simulation (EKS). We observe that FAs influence Aß dynamics distinctively in three broadly-defined FA concentration regimes containing non-micellar, pseudo-micellar or micellar phases. While the non-micellar phase promotes on-pathway fibrils, pseudo-micellar and micellar phases promote predominantly off-pathway oligomers, albeit via subtly different mechanisms. Importantly off-pathway oligomers saturate within a limited molecular size, and likely with a different overall conformation than those formed along the on-pathway, suggesting the generation of distinct conformeric strains of Aß, which may have profound phenotypic outcomes. Our results validate previous experimental observations and provide insights into potential influence of biological interfaces in modulating Aß aggregation pathways.

Determining causal miRNAs and their signaling cascade in diseases using an influence diffusion model.

In recent studies, miRNAs have been found to be extremely influential in many of the essential biological processes. They exhibit a self-regulatory mechanism through which they act as positive/negative regulators of expression of genes and other miRNAs. This has direct implications in the regulation of various pathophysiological conditions, signaling pathways and different types of cancers. Studying miRNA-disease associations has been an extensive area of research; however deciphering miRNA-miRNA network regulatory patterns in several diseases remains a challenge. In this study, we use information diffusion theory to quantify the influence diffusion in a miRNA-miRNA regulation network across multiple disease categories. Our proposed methodology determines the critical disease specific miRNAs which play a causal role in their signaling cascade and hence may regulate disease progression. We extensively validate our framework using existing computational tools from the literature. Furthermore, we implement our framework on a comprehensive miRNA expression data set for alcohol dependence and identify the causal miRNAs for alcohol-dependency in patients which were validated by the phase-shift in their expression scores towards the early stages of the disease. Finally, our computational framework for identifying causal miRNAs implicated in diseases is available as a free online tool for the greater scientific community.

Strain-specific Fibril Propagation by an Aß Dodecamer.

Low molecular weight oligomers of amyloid-ß (Aß) have emerged as the primary toxic agents in the etiology of Alzheimer disease (AD). Polymorphism observed within the aggregation end products of fibrils are known to arise due to microstructural differences among the oligomers. Diversity in aggregate morphology correlates with the differences in AD, cementing the idea that conformational strains of oligomers could be significant in phenotypic outcomes. Therefore, it is imperative to determine the ability of strains to faithfully propagate their structure. Here we report fibril propagation of an Aß42 dodecamer called large fatty acid-derived oligomers (LFAOs). The LFAO oligomeric strain selectively induces acute cerebral amyloid angiopathy (CAA) in neonatally-injected transgenic CRND8 mice. Propagation in-vitro occurs as a three-step process involving the association of LFAO units. LFAO-seeded fibrils possess distinct morphology made of repeating LFAO units that could be regenerated upon sonication. Overall, these data bring forth an important mechanistic perspective into strain-specific propagation of oligomers that has remained elusive thus far.