Identifying accurate biomarkers of cognitive decline is essential for advancing early diagnosis and prevention therapies in Alzheimer’s Disease. The Alzheimer’s Disease DREAM Challenge was designed as a computational crowdsourced project to benchmark the current state-of-the-art in predicting cognitive outcomes in Alzheimer’s Disease based on high-dimensional, publicly available genetic and structural imaging data. This meta-analysis failed to identify a meaningful predictor developed from either data modality, suggesting that alternate approaches should be considered for to prediction of cognitive performance.

Alzheimer’s Disease (AD) is among the most frequent neuro-degenerative diseases. Early diagnosis is essential for successful disease management and chance to attenuate symptoms by disease modifying drugs. In the past, a number of cerebrospinal fluid (CSF), plasma and neuro-imaging based biomarkers have been proposed. Still, in current clinical practice, AD diagnosis cannot be made until the patient shows clear signs of cognitive decline, which can partially be attributed to the multi-factorial nature of AD. In this work, we integrated genotype information, neuro-imaging as well as clinical data (including neuro-psychological measures) from ~900 normal and mild cognitively impaired (MCI) individuals and developed a highly accurate machine learning model to predict the time until AD is diagnosed. We performed an in-depth investigation of the relevant baseline characteristics that contributed to the AD risk prediction. More specifically, we used Bayesian Networks to uncover the interplay across biological scales between neuro-psychological assessment scores, single genetic variants, pathways and neuro-imaging related features. Together with information extracted from the literature, this allowed us to partially reconstruct biological mechanisms that could play a role in the conversion of normal/MCI into AD pathology. This in turn may open the door to novel therapeutic options in the future.

MotivationThe concept of a ‘mechanism-based taxonomy of human disease’ is currently replacing the outdated paradigm of diseases classified by clinical appearance. We have tackled the paradigm of mechanism-based patient subgroup identification in the challenging area of research on neurodegenerative diseases.ResultsWe have developed a knowledge base representing essential pathophysiology mechanisms of neurodegenerative diseases. Together with dedicated algorithms, this knowledge base forms the basis for a ‘mechanism-enrichment server’ that supports the mechanistic interpretation of multiscale, multimodal clinical data.Availability and implementationNeuroMMSig is available at http://neurommsig.scai.fraunhofer.de/Supplementary information Supplementary data are available at Bioinformatics online.

Perturbance in inflammatory pathways have been identified as one of the major factors which leads to neurodegenerative diseases (NDD). Owing to the limited access of human brain tissues and the immense complexity of the brain, animal models, specifically mouse models, play a key role in advancing the NDD field. However, many of these mouse models fail to reproduce the clinical manifestations and end points of the disease. NDD drugs, which passed the efficacy test in mice, were repeatedly not successful in clinical trials. There are numerous studies which are supporting and opposing the applicability of mouse models in neuroinflammation and NDD. In this paper, we assessed to what extend a mouse can mimic the cellular and molecular interactions in humans at a mechanism level. Based on our mechanistic modeling approach, we investigate the failure of a neuroinflammation targeted drug in the late phases of clinical trials based on the comparative analyses between the two species.

Ontologies and terminologies are used for interoperability of knowledge and data in a standard manner among interdisciplinary research groups. Existing imaging ontologies capture general aspects of the imaging domain as a whole such as methodological concepts or calibrations of imaging instruments. However, none of the existing ontologies covers the diagnostic features measured by imaging technologies in the context of neurodegenerative diseases. Therefore, the Neuro-Imaging Feature Terminology (NIFT) was developed to organize the knowledge domain of measured brain features in association with neurodegenerative diseases by imaging technologies. The purpose is to identify quantitative imaging biomarkers that can be extracted from multi-modal brain imaging data. This terminology attempts to cover measured features and parameters in brain scans relevant to disease progression. In this paper, we demonstrate the systematic retrieval of measured indices from literature and how the extracted knowledge can be further used for disease modeling that integrates neuroimaging features with molecular processes.

BackgroundNeurodegenerative diseases are incurable and debilitating indications with huge social and economic impact, where much is still to be learnt about the underlying molecular events. Mechanistic disease models could offer a knowledge framework to help decipher the complex interactions that occur at molecular and cellular levels. This motivates the need for the development of an approach integrating highly curated and heterogeneous data into a disease model of different regulatory data layers. Although several disease models exist, they often do not consider the quality of underlying data. Moreover, even with the current advancements in semantic web technology, we still do not have cure for complex diseases like Alzheimer’s disease. One of the key reasons accountable for this could be the increasing gap between generated data and the derived knowledge.ResultsIn this paper, we describe an approach, called as NeuroRDF, to develop an integrative framework for modeling curated knowledge in the area of complex neurodegenerative diseases. The core of this strategy lies in the usage of well curated and context specific data for integration into one single semantic web-based framework, RDF. This increases the probability of the derived knowledge to be novel and reliable in a specific disease context. This infrastructure integrates highly curated data from databases (Bind, IntAct, etc.), literature (PubMed), and gene expression resources (such as GEO and ArrayExpress). We illustrate the effectiveness of our approach by asking real-world biomedical questions that link these resources to prioritize the plausible biomarker candidates. Among the 13 prioritized candidate genes, we identified MIF to be a potential emerging candidate due to its role as a pro-inflammatory cytokine. We additionally report on the effort and challenges faced during generation of such an indication-specific knowledge base comprising of curated and quality-controlled data.ConclusionAlthough many alternative approaches have been proposed and practiced for modeling diseases, the semantic web technology is a flexible and well established solution for harmonized aggregation. The benefit of this work, to use high quality and context specific data, becomes apparent in speculating previously unattended biomarker candidates around a well-known mechanism, further leveraged for experimental investigations.Electronic supplementary materialThe online version of this article (doi:10.1186/s13326-016-0079-8) contains supplementary material, which is available to authorized users.

Molecular signaling pathways have been long used to demonstrate interactions among upstream causal molecules and downstream biological effects. They show the signal flow between cell compartments, the majority of which are represented as cartoons. These are often drawn manually by scanning through the literature, which is time-consuming, static, and non-interoperable. Moreover, these pathways are often devoid of context (condition and tissue) and biased toward certain disease conditions. Mining the scientific literature creates new possibilities to retrieve pathway information at higher contextual resolution and specificity. To address this challenge, we have created a pathway terminology system by combining signaling pathways and biological events to ensure a broad coverage of the entire pathway knowledge domain. This terminology was applied to mining biomedical papers and patents about neurodegenerative diseases with focus on Alzheimer's disease. We demonstrate the power of our approach by mapping literature-derived signaling pathways onto their corresponding anatomical regions in the human brain under healthy and Alzheimer's disease states. We demonstrate how this knowledge resource can be used to identify a putative mechanism explaining the mode-of-action of the approved drug Rasagiline, and show how this resource can be used for fingerprinting patents to support the discovery of pathway knowledge for Alzheimer's disease. Finally, we propose that based on next-generation cause-and-effect pathway models, a dedicated inventory of computer-processable pathway models specific to neurodegenerative diseases can be established, which hopefully accelerates context-specific enrichment analysis of experimental data with higher resolution and richer annotations.

Background: Neuroimaging markers provide quantitative insight into brain structure and function in neurodegenerative diseases, such as Alzheimer’s disease, where we lack mechanistic insights to explain pathophysiology. These mechanisms are often mediated by genes and genetic variations and are often studied through the lens of genome-wide association studies. Linking these two disparate layers (i.e., imaging and genetic variation) through causal relationships between biological entities involved in the disease’s etiology would pave the way to large-scale mechanistic reasoning and interpretation. Objective: We explore how genetic variants may lead to functional alterations of intermediate molecular traits, which can further impact neuroimaging hallmarks over a series of biological processes across multiple scales. Methods: We present an approach in which knowledge pertaining to single nucleotide polymorphism (SNPs) and imaging readouts is extracted from the literature, encoded in Biological Expression Language, and used in a novel workflow to assist in the functional interpretation of SNPs in a clinical context. Results: We demonstrate our approach in a case scenario which proposes KANSL1 as a candidate gene that accounts for the clinically reported correlation between the incidence of the genetic variants and hippocampal atrophy. We find that the workflow prioritizes multiple mechanisms reported in the literature through which KANSL1 may have an impact on hippocampal atrophy such as through the dysregulation of cell proliferation, synaptic plasticity, and metabolic processes. Conclusion: We have presented an approach that enables pinpointing relevant genetic variants as well as investigating their functional role in biological processes spanning across several, diverse biological scales.