Machine learning on longitudinal multi-modal data enables the understanding and prognosis of Alzheimer's disease progression.

Alzheimer's disease (AD) is a complex pathophysiological disease. Allowing for heterogeneity, not only in disease manifestations but also in different progression patterns, is critical for developing effective disease models that can be used in clinical and research settings. We introduce a machine learning model for identifying underlying patterns in Alzheimer's disease (AD) trajectory using longitudinal multi-modal data from the ADNI cohort and the AIBL cohort. Ten biologically and clinically meaningful disease-related states were identified from data, which constitute three non-overlapping stages (i.e., neocortical atrophy [NCA], medial temporal atrophy [MTA], and whole brain atrophy [WBA]) and two distinct disease progression patterns (i.e., NCA → WBA and MTA → WBA). The index of disease-related states provided a remarkable performance in predicting the time to conversion to AD dementia (C-Index: 0.923 ± 0.007). Our model shows potential for promoting the understanding of heterogeneous disease progression and early predicting the conversion time to AD dementia.

Identifying underlying patterns in Alzheimer's disease trajectory: a deep learning approach and Mendelian randomization analysis.

Background: Alzheimer's disease (AD) is a heterogeneously progressive neurodegeneration disorder with varied rates of deterioration, either between subjects or within different stages of a certain subject. Estimating the course of AD at early stages has treatment implications. We aimed to analyze disease progression to identify distinct patterns in AD trajectory. Methods: We proposed a deep learning model to identify underlying patterns in the trajectory from cognitively normal (CN) to a state of mild cognitive impairment (MCI) to AD dementia, by jointly predicting time-to-conversion and clustering out distinct subgroups characterized by comprehensive features as well as varied progression rates. We designed and validated our model on the ADNI dataset (1370 participants). Prediction of time-to-conversion in AD trajectory was used to validate the expression of the identified patterns. Causality between patterns and time-to-conversion was further inferred using Mendelian randomization (MR) analysis. External validation was performed on the AIBL dataset (233 participants). Findings: The proposed model clustered out patterns characterized by significantly different biomarkers and varied progression rates. The discovered patterns also showed a strong prediction ability, as indicated by hazard ratio (CN→MCI, HR = 3.51, p < 0.001; MCI→AD, HR = 8.11, p < 0.001), C-Index (CN→MCI, 0.618; MCI→AD, 0.718), and AUC (CN→MCI, 3 years 0.802, 5 years 0.876; MCI→AD, 3 years 0.914, 5 years 0.957). In the external validation cohort, our model demonstrated competitive performance on conversion time prediction (CN→MCI, C-Index = 0.693; MCI→AD, C-Index = 0.752). Moreover, suggestive associations between CN→MCI/MCI→AD patterns with four/three SNPs were mediated and MR analysis indicated a causal link between MCI→AD patterns and time-to-conversion in the first three years. Interpretation: Our proposed model identifies biologically and clinically meaningful patterns from real-world data and provides promising performance on time-to-conversion prediction in AD trajectory, which could promote the understanding of disease progression, facilitate clinical trial design, and provide potential for decision-making. Funding: The National Key Research and Development Program of China, the Key R&D Program of Zhejiang, and the National Nature Science Foundation of China.

A methodology for classifying tissue-specific metabolic and inflammatory receptor functions applied to subcutaneous and visceral adipose.

To achieve homeostasis, the human biological system relies on the interaction between organs through the binding of ligands secreted from source organs to receptors located on destination organs. Currently, the changing roles that receptors perform in tissues are only partially understood. Recently, a methodology based on receptor co-expression patterns to classify their tissue-specific metabolic functions was suggested. Here we present an advanced framework to predict an additional class of inflammatory receptors that use a feature space of biological pathway enrichment analysis scores of co-expression networks and their eigengene correlations. These are fed into three machine learning classifiers-eXtreme Gradient Boosting (XGBoost), Support Vector Machines (SVM), and K-Nearest Neighbors (k-NN). We applied our methodology to subcutaneous and visceral adipose gene expression datasets derived from the GTEx (Genotype-Tissue Expression) project and compared the predictions. The XGBoost model demonstrated the best performance in predicting the pre-labeled receptors, with an accuracy of 0.89/0.8 in subcutaneous/visceral adipose. We analyzed ~700 receptors to predict eight new metabolic and 15 new inflammatory functions of receptors and four new metabolic functions for known inflammatory receptors in both adipose tissues. We cross-referenced multiple predictions using the published literature. Our results establish a picture of the changing functions of receptors for two adipose tissues that can be beneficial for drug development.

Predicting mechanical ventilation effects on six human tissue transcriptomes.

BACKGROUND: Mechanical ventilation (MV) is a lifesaving therapy used for patients with respiratory failure. Nevertheless, MV is associated with numerous complications and increased mortality. The aim of this study is to define the effects of MV on gene expression of direct and peripheral human tissues. METHODS: Classification models were applied to Genotype-Tissue Expression Project (GTEx) gene expression data of six representative tissues-liver, adipose, skin, nerve-tibial, muscle and lung, for performance comparison and feature analysis. We utilized 18 prediction models using the Random Forest (RF), XGBoost (eXtreme Gradient Boosting) decision tree and ANN (Artificial Neural Network) methods to classify ventilation and non-ventilation samples and to compare their prediction performance for the six tissues. In the model comparison, the AUC (area under receiver operating curve), accuracy, precision, recall, and F1 score were used to evaluate the predictive performance of each model. We then conducted feature analysis per each tissue to detect MV marker genes followed by pathway enrichment analysis for these genes. RESULTS: XGBoost outperformed the other methods and predicted samples had undergone MV with an average accuracy for the six tissues of 0.951 and average AUC of 0.945. The feature analysis detected a combination of MV marker genes per each tested tissue, some common across several tissues. MV marker genes were mainly related to inflammation and fibrosis as well as cell development and movement regulation. The MV marker genes were significantly enriched in inflammatory and viral pathways. CONCLUSION: The XGBoost method demonstrated clear enhanced performance and feature analysis compared to the other models. XGBoost was helpful in detecting the tissue-specific marker genes for identifying transcriptomic changes related to MV. Our results show that MV is associated with reduced development and movement in the tissues and higher inflammation and injury not only in direct tissues such as the lungs but also in peripheral tissues and thus should be carefully considered before being implemented.

Multimodal bioinformatic analyses of the neurodegenerative disease-associated TECPR2 gene reveal its diverse roles.

BACKGROUND: Loss of tectonin ß-propeller repeat-containing 2 (TECPR2) function has been implicated in an array of neurodegenerative disorders, yet its physiological function remains largely unknown. Understanding TECPR2 function is essential for developing much needed precision therapeutics for TECPR2-related diseases. METHODS: We leveraged considerable amounts of functional data to obtain a comprehensive perspective of the role of TECPR2 in health and disease. We integrated expression patterns, population variation, phylogenetic profiling, protein-protein interactions and regulatory network data for a minimally biased multimodal functional analysis. Genes and proteins linked to TECPR2 via multiple lines of evidence were subject to functional enrichment analyses to identify molecular mechanisms involving TECPR2. RESULTS: TECPR2 was found to be part of a tight neurodevelopmental gene expression programme that includes KIF1A, ATXN1, TOM1L2 and FA2H, all implicated in neurological diseases. Functional enrichment analyses of TECPR2-related genes converged on a role in late autophagy and ribosomal processes. Large-scale population variation data demonstrated that this role is non-redundant. CONCLUSIONS: TECPR2 might serve as an indicator for the energy balance between protein synthesis and autophagy, and a marker for diseases associated with their imbalance, such as Alzheimer's disease and Huntington's disease. Specifically, we speculate that TECPR2 plays an important role as a proteostasis regulator during synaptogenesis, highlighting its importance in developing neurons. By advancing our understanding of TECPR2 function, this work provides an essential stepping stone towards the development of precision diagnostics and targeted treatment options for TECPR2-related disorders.

Model-based pathway enrichment analysis applied to the TGF-beta regulation of autophagy in autism.

To differentiate between conditions of health and disease, current pathway enrichment analysis methods detect the differential expression of distinct biological pathways. System-level model-driven approaches, however, are lacking. Here we present a new methodology that uses a dynamic model to suggest a unified subsystem to better differentiate between diseased and healthy conditions. Our methodology includes the following steps: 1) detecting connections between relevant differentially expressed pathways; 2) construction of a unified in silico model, a stochastic Petri net model that links these distinct pathways; 3) model execution to predict subsystem activation; and 4) enrichment analysis of the predicted subsystem. We apply our approach to the TGF-beta regulation of the autophagy system implicated in autism. Our model was constructed manually, based on the literature, to predict, using model simulation, the TGF-beta-to-autophagy active subsystem and downstream gene expression changes associated with TGF-beta, which go beyond the individual findings derived from literature. We evaluated the in silico predicted subsystem and found it to be co-expressed in the normative whole blood human gene expression data. Finally, we show our subsystem's gene set to be significantly differentially expressed in two independent datasets of blood samples of ASD (autistic spectrum disorders) individuals as opposed to controls. Our study demonstrates that dynamic pathway unification can define a new refined subsystem that can significantly differentiate between disease conditions.

A methodology for predicting tissue-specific metabolic roles of receptors applied to subcutaneous adipose.

The human biological system uses 'inter-organ' communication to achieve a state of homeostasis. This communication occurs through the response of receptors, located on target organs, to the binding of secreted ligands from source organs. Albeit years of research, the roles these receptors play in tissues is only partially understood. This work presents a new methodology based on the enrichment analysis scores of co-expression networks fed into support vector machines (SVMs) and k-NN classifiers to predict the tissue-specific metabolic roles of receptors. The approach is primarily based on the detection of coordination patterns of receptors expression. These patterns and the enrichment analysis scores of their co-expression networks were used to analyse ~ 700 receptors and predict metabolic roles of receptors in subcutaneous adipose. To facilitate supervised learning, a list of known metabolic and non-metabolic receptors was constructed using a semi-supervised approach following literature-based verification. Our approach confirms that pathway enrichment scores are good signatures for correctly classifying the metabolic receptors in adipose. We also show that the k-NN method outperforms the SVM method in classifying metabolic receptors. Finally, we predict novel metabolic roles of receptors. These predictions can enhance biological understanding and the development of new receptor-targeting metabolic drugs.

Batch correction evaluation framework using a-priori gene-gene associations: applied to the GTEx dataset.

BACKGROUND: Correcting a heterogeneous dataset that presents artefacts from several confounders is often an essential bioinformatics task. Attempting to remove these batch effects will result in some biologically meaningful signals being lost. Thus, a central challenge is assessing if the removal of unwanted technical variation harms the biological signal that is of interest to the researcher. RESULTS: We describe a novel framework, B-CeF, to evaluate the effectiveness of batch correction methods and their tendency toward over or under correction. The approach is based on comparing co-expression of adjusted gene-gene pairs to a-priori knowledge of highly confident gene-gene associations based on thousands of unrelated experiments derived from an external reference. Our framework includes three steps: (1) data adjustment with the desired methods (2) calculating gene-gene co-expression measurements for adjusted datasets (3) evaluating the performance of the co-expression measurements against a gold standard. Using the framework, we evaluated five batch correction methods applied to RNA-seq data of six representative tissue datasets derived from the GTEx project. CONCLUSIONS: Our framework enables the evaluation of batch correction methods to better preserve the original biological signal. We show that using a multiple linear regression model to correct for known confounders outperforms factor analysis-based methods that estimate hidden confounders. The code is publicly available as an R package.

Ulcerative colitis mucosal transcriptomes reveal mitochondriopathy and personalized mechanisms underlying disease severity and treatment response.

Molecular mechanisms driving disease course and response to therapy in ulcerative colitis (UC) are not well understood. Here, we use RNAseq to define pre-treatment rectal gene expression, and fecal microbiota profiles, in 206 pediatric UC patients receiving standardised therapy. We validate our key findings in adult and paediatric UC cohorts of 408 participants. We observe a marked suppression of mitochondrial genes and function across cohorts in active UC, and that increasing disease severity is notable for enrichment of adenoma/adenocarcinoma and innate immune genes. A subset of severity genes improves prediction of corticosteroid-induced remission in the discovery cohort; this gene signature is also associated with response to anti-TNFα and anti-α4ß7 integrin in adults. The severity and therapeutic response gene signatures were in turn associated with shifts in microbes previously implicated in mucosal homeostasis. Our data provide insights into UC pathogenesis, and may prioritise future therapies for nonresponders to current approaches.

A model-driven methodology for exploring complex disease comorbidities applied to autism spectrum disorder and inflammatory bowel disease.

We propose a model-driven methodology aimed to shed light on complex disorders. Our approach enables exploring shared etiologies of comorbid diseases at the molecular pathway level. The method, Comparative Comorbidities Simulation (CCS), uses stochastic Petri net simulation for examining the phenotypic effects of perturbation of a network known to be involved in comorbidities to predict new roles for mutations in comorbid conditions. To demonstrate the utility of our novel methodology, we investigated the molecular convergence of autism spectrum disorder (ASD) and inflammatory bowel disease (IBD) on the autophagy pathway. In addition to validation by domain experts, we used formal analyses to demonstrate the model's self-consistency. We then used CCS to compare the effects of loss of function (LoF) mutations previously implicated in either ASD or IBD on the autophagy pathway. CCS identified similar dynamic consequences of these mutations in the autophagy pathway. Our method suggests that two LoF mutations previously implicated in IBD may contribute to ASD, and one ASD-implicated LoF mutation may play a role in IBD. Future targeted genomic or functional studies could be designed to directly test these predictions.

Conceptual modeling of mRNA decay provokes new hypotheses.

Biologists are required to integrate large amounts of data to construct a working model of the system under investigation. This model is often informal and stored mentally or textually, making it prone to contain undetected inconsistencies, inaccuracies, or even contradictions, not much less than a representation in free natural language. Using Object-Process Methodology (OPM), a formal yet visual and humanly accessible conceptual modeling language, we have created an executable working model of the mRNA decay process in Saccharomyces cerevisiae, as well as the import of its components to the nucleus following mRNA decay. We show how our model, which incorporates knowledge from 43 articles, can reproduce outcomes that match the experimental findings, evaluate hypotheses, and predict new possible outcomes. Moreover, we were able to analyze the effects of the mRNA decay model perturbations related to gene and interaction deletions, and predict the nuclear import of certain decay factors, which we then verified experimentally. In particular, we verified experimentally the hypothesis that Rpb4p, Lsm1p, and Pan2p remain bound to the RNA 3'-untranslated region during the entire process of the 5' to 3' degradation of the RNA open reading frame. The model has also highlighted erroneous hypotheses that indeed were not in line with the experimental outcomes. Beyond the scientific value of these specific findings, this work demonstrates the value of the conceptual model as an in silico vehicle for hypotheses generation and testing, which can reinforce, and often even replace, risky, costlier wet lab experiments.

Conceptual Model-based Systems Biology: mapping knowledge and discovering gaps in the mRNA transcription cycle.

We propose a Conceptual Model-based Systems Biology framework for qualitative modeling, executing, and eliciting knowledge gaps in molecular biology systems. The framework is an adaptation of Object-Process Methodology (OPM), a graphical and textual executable modeling language. OPM enables concurrent representation of the system's structure-the objects that comprise the system, and behavior-how processes transform objects over time. Applying a top-down approach of recursively zooming into processes, we model a case in point-the mRNA transcription cycle. Starting with this high level cell function, we model increasingly detailed processes along with participating objects. Our modeling approach is capable of modeling molecular processes such as complex formation, localization and trafficking, molecular binding, enzymatic stimulation, and environmental intervention. At the lowest level, similar to the Gene Ontology, all biological processes boil down to three basic molecular functions: catalysis, binding/dissociation, and transporting. During modeling and execution of the mRNA transcription model, we discovered knowledge gaps, which we present and classify into various types. We also show how model execution enhances a coherent model construction. Identification and pinpointing knowledge gaps is an important feature of the framework, as it suggests where research should focus and whether conjectures about uncertain mechanisms fit into the already verified model.

A methodology for eliciting and modeling exceptions.

Exceptions in safety-critical systems must be addressed during conceptual design and risk analysis. We developed a conceptual model of exceptions, a methodology for eliciting and modeling exceptions, and templates for modeling them in an extension of the Object-Process Methodology (OPM)-a system analysis and design methodology and language that uses a single graphical model for describing systems, including their timing exceptions, which has been shown to be an effective modeling methodology. Using an antibiotics treatment guideline as a case study, we demonstrate the value of our approach in eliciting and modeling exceptions that occur in clinical care systems.