Integrating Clinical Data and Imputed Transcriptome from GWAS to Uncover Complex Disease Subtypes: Applications in Psychiatry and Cardiology

Leveraging Deep Learning for Fine-Grained Categorization of Parkinson's Disease Progression Levels through Analysis of Vocal Acoustic Patterns

Speech impairments often emerge as one of the primary indicators of Parkinson's disease (PD), albeit not readily apparent in its early stages. While previous studies focused predominantly on binary PD detection, this research explored the use of deep learning models to automatically classify sustained vowel recordings into healthy controls, mild PD, or severe PD based on motor symptom severity scores. Popular convolutional neural network (CNN) architectures, VGG and ResNet, as well as vision transformers, Swin, were fine-tuned on log mel spectrogram image representations of the segmented voice data. Furthermore, the research investigated the effects of audio segment lengths and specific vowel sounds on the performance of these models. The findings indicated that implementing longer segments yielded better performance. The models showed strong capability in distinguishing PD from healthy subjects, achieving over 95% precision. However, reliably discriminating between mild and severe PD cases remained challenging. The VGG16 achieved the best overall classification performance with 91.8% accuracy and the largest area under the ROC curve. Furthermore, focusing analysis on the vowel /u/ could further improve accuracy to 96%. Applying visualization techniques like Grad-CAM also highlighted how CNN models focused on localized spectrogram regions while transformers attended to more widespread patterns. Overall, this work showed the potential of deep learning for non-invasive screening and monitoring of PD progression from voice recordings, but larger multi-class labeled datasets are needed to further improve severity classification.

Genetic imputation of kidney transcriptome, proteome and multi-omics illuminates new blood pressure and hypertension targets

Genetic mechanisms of blood pressure (BP) regulation remain poorly defined. Using kidney-specific epigenomic annotations and 3D genome information we generated and validated gene expression prediction models for the purpose of transcriptome-wide association studies in 700 human kidneys. We identified 889 kidney genes associated with BP of which 399 were prioritised as contributors to BP regulation. Imputation of kidney proteome and microRNAome uncovered 97 renal proteins and 11 miRNAs associated with BP. Integration with plasma proteomics and metabolomics illuminated circulating levels of myo-inositol, 4-guanidinobutanoate and angiotensinogen as downstream effectors of several kidney BP genes (SLC5A11, AGMAT, AGT, respectively). We showed that genetically determined reduction in renal expression may mimic the effects of rare loss-of-function variants on kidney mRNA/protein and lead to an increase in BP (e.g., ENPEP). We demonstrated a strong correlation (r = 0.81) in expression of protein-coding genes between cells harvested from urine and the kidney highlighting a diagnostic potential of urinary cell transcriptomics. We uncovered adenylyl cyclase activators as a repurposing opportunity for hypertension and illustrated examples of BP-elevating effects of anticancer drugs (e.g. tubulin polymerisation inhibitors). Collectively, our studies provide new biological insights into genetic regulation of BP with potential to drive clinical translation in hypertension.

Unraveling patient heterogeneity in complex diseases through individualized co-expression networks: a perspective

This perspective highlights the potential of individualized networks as a novel strategy for studying complex diseases through patient stratification, enabling advancements in precision medicine. We emphasize the impact of interpatient heterogeneity resulting from genetic and environmental factors and discuss how individualized networks improve our ability to develop treatments and enhance diagnostics. Integrating system biology, combining multimodal information such as genomic and clinical data has reached a tipping point, allowing the inference of biological networks at a single-individual resolution. This approach generates a specific biological network per sample, representing the individual from which the sample originated. The availability of individualized networks enables applications in personalized medicine, such as identifying malfunctions and selecting tailored treatments. In essence, reliable, individualized networks can expedite research progress in understanding drug response variability by modeling heterogeneity among individuals and enabling the personalized selection of pharmacological targets for treatment. Therefore, developing diverse and cost-effective approaches for generating these networks is crucial for widespread application in clinical services.

The genetic basis of onset age in schizophrenia: evidence and models

Schizophrenia is a heritable neurocognitive disorder affecting about 1% of the population, and usually has an onset age at around 21-25 in males and 25-30 in females. Recent advances in genetics have helped to identify many common and rare variants for the liability to schizophrenia. Earlier evidence appeared to suggest that younger onset age is associated with higher genetic liability to schizophrenia. Clinical longitudinal research also found that early and very-early onset schizophrenia are associated with poor clinical, neurocognitive, and functional profiles. A recent study reported a heritability of 0.33 for schizophrenia onset age, but the genetic basis of this trait in schizophrenia remains elusive. In the pre-Genome-Wide Association Study (GWAS) era, genetic loci found to be associated with onset age were seldom replicated. In the post-Genome-Wide Association Study era, new conceptual frameworks are needed to clarify the role of onset age in genetic research in schizophrenia, and to identify its genetic basis. In this review, we first discussed the potential of onset age as a characterizing/subtyping feature for psychosis, and as an important phenotypic dimension of schizophrenia. Second, we reviewed the methods, samples, findings and limitations of previous genetic research on onset age in schizophrenia. Third, we discussed a potential conceptual framework for studying the genetic basis of onset age, as well as the concepts of susceptibility, modifier, and "mixed" genes. Fourth, we discussed the limitations of this review. Lastly, we discussed the potential clinical implications for genetic research of onset age of schizophrenia, and how future research can unveil the potential mechanisms for this trait.

NetMoST: A network-based machine learning approach for subtyping schizophrenia using polygenic SNP allele biomarkers

Subtyping neuropsychiatric disorders like schizophrenia is essential for improving the diagnosis and treatment of complex diseases. Subtyping schizophrenia is challenging because it is polygenic and genetically heterogeneous, rendering the standard symptom-based diagnosis often unreliable and unrepeatable. We developed a novel network-based machine-learning approach, netMoST, to subtyping psychiatric disorders. NetMoST identifies polygenic risk SNP-allele modules from genome-wide genotyping data as polygenic haplotype biomarkers (PHBs) for disease subtyping. We applied netMoST to subtype a cohort of schizophrenia subjects into three distinct biotypes with differentiable genetic, neuroimaging and functional characteristics. The PHBs of the first biotype (36.9% of all patients) were related to neurodevelopment and cognition, the PHBs of the second biotype (28.4%) were enriched for neuroimmune functions, and the PHBs of the third biotype (34.7%) were associated with the transport of calcium ions and neurotransmitters. Neuroimaging patterns provided additional support to the new biotypes, with unique regional homogeneity (ReHo) patterns observed in the brains of each biotype compared with healthy controls. Our findings demonstrated netMoST's capability for uncovering novel biotypes of complex diseases such as schizophrenia. The results also showed the power of exploring polygenic allelic patterns that transcend the conventional GWAS approaches.

Integrative omics of schizophrenia: from genetic determinants to clinical classification and risk prediction

Schizophrenia (SCZ) is a debilitating neuropsychiatric disorder with high heritability and complex inheritance. In the past decade, successful identification of numerous susceptibility loci has provided useful insights into the molecular etiology of SCZ. However, applications of these findings to clinical classification and diagnosis, risk prediction, or intervention for SCZ have been limited, and elucidating the underlying genomic and molecular mechanisms of SCZ is still challenging. More recently, multiple Omics technologies - genomics, transcriptomics, epigenomics, proteomics, metabolomics, connectomics, and gut microbiomics - have all been applied to examine different aspects of SCZ pathogenesis. Integration of multi-Omics data has thus emerged as an approach to provide a more comprehensive view of biological complexity, which is vital to enable translation into assessments and interventions of clinical benefit to individuals with SCZ. In this review, we provide a broad survey of the single-omics studies of SCZ, summarize the advantages and challenges of different Omics technologies, and then focus on studies in which multiple omics data are integrated to unravel the complex pathophysiology of SCZ. We believe that integration of multi-Omics technologies would provide a roadmap to create a more comprehensive picture of interactions involved in the complex pathogenesis of SCZ, constitute a rich resource for elucidating the potential molecular mechanisms of the illness, and eventually improve clinical assessments and interventions of SCZ to address clinical translational questions from bench to bedside.

Shaping the Trans-Scale Properties of Schizophrenia via Cerebral Alterations on Magnetic Resonance Imaging and Single-Nucleotide Polymorphisms of Coding and Non-Coding Regions

Schizophrenia is a complex mental illness with genetic heterogeneity, which is often accompanied by alterations in brain structure and function. The neurobiological mechanism of schizophrenia associated with heredity remains unknown. Recently, the development of trans-scale and multi-omics methods that integrate gene and imaging information sheds new light on the nature of schizophrenia. In this article, we summarized the results of brain structural and functional changes related to the specific single-nucleotide polymorphisms (SNPs) in the past decade, and the SNPs were divided into non-coding regions and coding regions, respectively. It is hoped that the relationship between SNPs and cerebral alterations can be displayed more clearly and intuitively, so as to provide fresh approaches for the discovery of potential biomarkers and the development of clinical accurate individualized treatment decision-making.

A framework to decipher the genetic architecture of combinations of complex diseases: Applications in cardiovascular medicine

MOTIVATION

Currently, most genome-wide association studies (GWAS) are studies of a single disease against controls. However, an individual is often affected by more than one condition. For example, coronary artery disease (CAD) is often comorbid with type 2 diabetes (T2DM). Similarly, it is clinically meaningful to study patients with one disease but without a related comorbidity. For example, obese T2DM may have different pathophysiology from non-obese T2DM.

RESULTS

We developed a statistical framework (CombGWAS) to uncover susceptibility variants for comorbid disorders (or a disorder without comorbidity), using GWAS summary statistics only. In essence, we mimicked a case-control GWAS in which the cases are affected with comorbidities or a disease without comorbidity. We extended our methodology to analyze continuous traits with clinically meaningful categories (e.g. lipids), and combination of more than 2 traits.We verified the feasibility and validity of our method by applying it to simulated scenarios and four cardiometabolic (CM) traits. In total, we identified 384 and 587 genomic risk loci respectively for 6 comorbidities and 12 CM disease 'subtypes' without a relevant comorbidity. Genetic correlation analysis revealed that some subtypes may be biologically distinct from others. Further Mendelian randomization analysis showed differential causal effects of different subtypes to relevant complications. For example, we found that obese T2DM is causally related to increased risk of CAD (p=2.62E-11).

AVAILABILITY

The R code is available at: https://github.com/LiangyingYin/CombGWAS.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Machine learning analysis: general features, requirements and cardiovascular applications

Artificial intelligence represents the science which will probably change the future of medicine by solving actually challenging issues. In this special article, the general features of machine learning are discussed. First, a background explanation regarding the division of artificial intelligence, machine learning and deep learning is given and a focus on the structure of machine learning subgroups is shown. The traditional process of a machine learning analysis is described, starting from the collection of data, across features engineering, modelling and till the validation and deployment phase. Due to the several applications of machine learning performed in literature in the last decades and the lack of some guidelines, the need of a standardization for reporting machine learning analysis results emerged. Some possible standards for reporting machine learning results are identified and discussed deeply; these are related to study population (number of subjects), repeatability of the analysis, validation, results, comparison with current practice. The way to the use of machine learning in clinical practice is open and the hope is that, with emerging technology and advanced digital and computational tools, available from hospitalization and subsequently after discharge, it will also be possible, with the help of increasingly powerful hardware, to build assistance strategies useful in clinical practice.

Leveraging eQTLs to identify individual-level tissue of interest for a complex trait

Genetic predisposition for complex traits often acts through multiple tissues at different time points during development. As a simple example, the genetic predisposition for obesity could be manifested either through inherited variants that control metabolism through regulation of genes expressed in the brain, or that control fat storage through dysregulation of genes expressed in adipose tissue, or both. Here we describe a statistical approach that leverages tissue-specific expression quantitative trait loci (eQTLs) corresponding to tissue-specific genes to prioritize a relevant tissue underlying the genetic predisposition of a given individual for a complex trait. Unlike existing approaches that prioritize relevant tissues for the trait in the population, our approach probabilistically quantifies the tissue-wise genetic contribution to the trait for a given individual. We hypothesize that for a subgroup of individuals the genetic contribution to the trait can be mediated primarily through a specific tissue. Through simulations using the UK Biobank, we show that our approach can predict the relevant tissue accurately and can cluster individuals according to their tissue-specific genetic architecture. We analyze body mass index (BMI) and waist to hip ratio adjusted for BMI (WHRadjBMI) in the UK Biobank to identify subgroups of individuals whose genetic predisposition act primarily through brain versus adipose tissue, and adipose versus muscle tissue, respectively. Notably, we find that these individuals have specific phenotypic features beyond BMI and WHRadjBMI that distinguish them from random individuals in the data, suggesting biological effects of tissue-specific genetic contribution for these traits.

Artificial intelligence and machine learning-aided drug discovery in central nervous system diseases: State-of-the-arts and future directions

Neurological disorders significantly outnumber diseases in other therapeutic areas. However, developing drugs for central nervous system (CNS) disorders remains the most challenging area in drug discovery, accompanied with the long timelines and high attrition rates. With the rapid growth of biomedical data enabled by advanced experimental technologies, artificial intelligence (AI) and machine learning (ML) have emerged as an indispensable tool to draw meaningful insights and improve decision making in drug discovery. Thanks to the advancements in AI and ML algorithms, now the AI/ML-driven solutions have an unprecedented potential to accelerate the process of CNS drug discovery with better success rate. In this review, we comprehensively summarize AI/ML-powered pharmaceutical discovery efforts and their implementations in the CNS area. After introducing the AI/ML models as well as the conceptualization and data preparation, we outline the applications of AI/ML technologies to several key procedures in drug discovery, including target identification, compound screening, hit/lead generation and optimization, drug response and synergy prediction, de novo drug design, and drug repurposing. We review the current state-of-the-art of AI/ML-guided CNS drug discovery, focusing on blood-brain barrier permeability prediction and implementation into therapeutic discovery for neurological diseases. Finally, we discuss the major challenges and limitations of current approaches and possible future directions that may provide resolutions to these difficulties.

Prioritization of schizophrenia risk genes from GWAS results by integrating multi-omics data

Schizophrenia (SCZ) is a polygenic disease with a heritability approaching 80%. Over 100 SCZ-related loci have so far been identified by genome-wide association studies (GWAS). However, the risk genes associated with these loci often remain unknown. We present a new risk gene predictor, rGAT-omics, that integrates multi-omics data under a Bayesian framework by combining the Hotelling and Box-Cox transformations. The Bayesian framework was constructed using gene ontology, tissue-specific protein-protein networks, and multi-omics data including differentially expressed genes in SCZ and controls, distance from genes to the index single-nucleotide polymorphisms (SNPs), and de novo mutations. The application of rGAT-omics to the 108 loci identified by a recent GWAS study of SCZ predicted 103 high-risk genes (HRGs) that explain a high proportion of SCZ heritability (Enrichment = 43.44 and [Formula: see text]). HRGs were shown to be significantly ([Formula: see text]) enriched in genes associated with neurological activities, and more likely to be expressed in brain tissues and SCZ-associated cell types than background genes. The predicted HRGs included 16 novel genes not present in any existing databases of SCZ-associated genes or previously predicted to be SCZ risk genes by any other method. More importantly, 13 of these 16 genes were not the nearest to the index SNP markers, and them would have been difficult to identify as risk genes by conventional approaches while ten out of the 16 genes are associated with neurological functions that make them prime candidates for pathological involvement in SCZ. Therefore, rGAT-omics has revealed novel insights into the molecular mechanisms underlying SCZ and could provide potential clues to future therapies.

Präzisionspsychiatrie und der Beitrag von Brain Imaging und anderen Biomarkern

'Precision Psychiatry' as the psychiatric variant of 'Precision Medicine' aims to provide high-level diagnosis and treatment based on robust biomarkers and tailored to the individual clinical, neurobiological, and genetic constitution of the patient. The specific peculiarity of psychiatry, in which disease entities are normatively defined based on clinical experience and are also significantly influenced by contemporary history, society and philosophy, has so far made the search for valid and reliable psychobiological connections difficult. Nevertheless, considerable progress has now been made in all areas of psychiatric research, made possible above all by the critical review and renewal of previous concepts of disease and psychopathology, the increased orientation towards neurobiology and genetics, and in particular the use of machine learning methods. Notably, modern machine learning methods make it possible to integrate high-dimensional and multimodal data sets and generate models which provide new psychobiological insights and offer the possibility of individualized, biomarker-driven single-subject prediction of diagnosis, therapy response and prognosis. The aim of the present review is therefore to introduce the concept of 'Precision Psychiatry' to the interested reader, to concisely present modern, machine learning methods required for this, and to clearly present the current state and future of biomarker-based 'precision psychiatry'.

Approaches to Defining Common and Dissociable Neurobiological Deficits Associated With Psychopathology in Youth

Psychiatric disorders show high rates of comorbidity and nonspecificity of presenting clinical symptoms, while demonstrating substantial heterogeneity within diagnostic categories. Notably, many of these psychiatric disorders first manifest in youth. We review progress and next steps in efforts to parse heterogeneity in psychiatric symptoms in youths by identifying abnormalities within neural circuits. To address this fundamental challenge in psychiatry, a number of methods have been proposed. We provide an overview of these methods, broadly organized into dimensional versus categorical approaches and single-view versus multiview approaches. Dimensional approaches including factor analysis and canonical correlation analysis aim to capture dimensional associations between psychopathology and brain measures across a continuous spectrum from health to disease. In contrast, categorical approaches, such as clustering and community detection, aim to identify subtypes of individuals within a class of symptoms or brain features. We highlight several studies that apply these methods to samples of youths and discuss issues to consider when using these approaches. Finally, we end by highlighting avenues for future research.

Turning genome-wide association study findings into opportunities for drug repositioning

Drug development is a very costly and lengthy process, while repositioned or repurposed drugs could be brought into clinical practice within a shorter time-frame and at a much reduced cost. Numerous computational approaches to drug repositioning have been developed, but methods utilizing genome-wide association studies (GWASs) data are less explored.

The past decade has observed a massive growth in the amount of data from GWAS; the rich information contained in GWAS has great potential to guide drug repositioning or discovery. While multiple tools are available for finding the most relevant genes from GWAS hits, searching for top susceptibility genes is only one way to guide repositioning, which has its own limitations.

Here we provide a comprehensive review of different computational approaches that employ GWAS data to guide drug repositioning. These methods include selecting top candidate genes from GWAS as drug targets, deducing drug candidates based on drug-drug and disease-disease similarities, searching for reversed expression profiles between drugs and diseases, pathway-based methods as well as approaches based on analysis of biological networks. Each method is illustrated with examples, and their respective strengths and limitations are discussed. We also discussed several areas for future research.