Multimodal AI/ML for discovering novel biomarkers and predicting disease using multi-omics profiles of patients with cardiovascular diseases

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Nov. 3, 2024

Cardiovascular diseases (CVDs) are complex, multifactorial conditions that require personalized assessment and treatment. Advancements in multi-omics technologies, namely RNA sequencing whole-genome sequencing, have provided translational researchers with a comprehensive view of the human genome. The efficient synthesis analysis this data through integrated approach characterizes genetic variants alongside expression patterns linked to emerging phenotypes, can reveal novel biomarkers enable segmentation patient populations based on risk factors. In study, we present cutting-edge methodology rooted integration traditional bioinformatics, classical statistics, multimodal machine learning techniques. Our has potential uncover intricate mechanisms underlying CVD, enabling patient-specific response profiling. We sourced transcriptomic single nucleotide polymorphisms (SNPs) from both CVD patients healthy controls. By integrating these datasets clinical demographic information, generated profiles. Utilizing robust feature selection approach, identified signature 27 features SNPs effective predictors CVD. Differential analysis, combined minimum redundancy maximum relevance selection, highlighted explain disease phenotype. This prioritizes biological efficiency learning. employed Combination Annotation Dependent Depletion scores allele frequencies identify pathogenic characteristics patients. Classification models trained demonstrated high-accuracy predictions for best performing was an XGBoost classifier optimized via Bayesian hyperparameter tuning, which able correctly classify all our test dataset. Using SHapley Additive exPlanations, created assessments patients, offering further contextualization setting. Across cohort, RPL36AP37 HBA1 were scored as most important predicting CVDs. A literature review revealed substantial portion diagnostic previously been associated framework propose study is unbiased generalizable other disorders.

Language: Английский

---

Big data and transformative bioinformatics in genomic diagnostics and beyond

Parkinsonism & Related Disorders, Journal Year: 2025, Volume and Issue: unknown, P. 107311 - 107311

Published: Feb. 1, 2025

The current era of high-throughput analysis-driven research offers invaluable insights into disease etiologies, accurate diagnostics, pathogenesis, and personalized therapy. In the field movement disorders, investigators are facing an increasing growth in volume produced patient-derived datasets, providing substantial opportunities for precision medicine approaches based on extensive information accessibility advanced annotation practices. Integrating data from multiple sources, including phenomics, genomics, multi-omics, is crucial comprehensively understanding different types disorders. Here, we explore formats analytics big generated patients with strategies to meaningfully share optimized patient benefit. We review computational methods that essential accelerate process evaluating amounts specialized collected. Based concrete examples, highlight how bioinformatic facilitate translation multidimensional biological clinically relevant knowledge. Moreover, outline feasibility computer-aided therapeutic target evaluation, discuss importance expanding focus understudied phenotypes such as dystonia.

Language: Английский

---

Integrated bioinformatics analysis and biological experiments to identify key immune genes in vascular dementia

Frontiers in Immunology, Journal Year: 2025, Volume and Issue: 16

Published: March 24, 2025

Objectives This study aimed to identify key immune genes provide new perspectives on the mechanisms and diagnosis of vascular dementia (VaD) based bioinformatic methods combined with biological experiments in mice. Methods We obtained gene expression profiles from a Gene Expression Omnibus database (GSE186798). The data were analysed using integrated bioinformatics machine learning techniques pinpoint potential immune-related for diagnosing VaD. Moreover, diagnostic accuracy was evaluated through receiver operating characteristic curve analysis. microRNA, transcription factor (TF), drug-regulating hub predicted database. Immune cell infiltration has been studied investigate dysregulation cells patients To evaluate cognitive impairment, mice bilateral common carotid artery stenosis (BCAS) subjected behavioural tests 30 d after chronic cerebral hypoperfusion. BCAS determined quantitative polymerase chain reaction(qPCR). Results results set enrichment variation analyses indicated that pathways upregulated A total 1620 included dataset, 323 differentially expressed examined GSE186798 dataset. Thirteen identified differential Protein-protein interaction network design functional analysis performed system as main subject. value, two core selected learning. Two putative genes, Rac family small GTPase 1( RAC1 ) CKLF-like MARVEL transmembrane domain containing 5 ( CMTM5 exhibit good value. Their high confidence levels confirmed by validating each biomarker different According GeneMANIA, VaD pathophysiology is strongly associated inflammatory responses. used construct miRNA gene, TFs-hub drug-hub networks. Varying also observed. In animal experiments, mouse model employed mimic humans, further Morris water maze test. mRNA significantly reduced group, which consistent Conclusions are frontal lobes mice, suggesting their biomarkers prognosis These findings pave way exploring novel molecular at preventing or treating

Language: Английский

---

AOPEP-related autosomal recessive dystonia: update on Zech-Boesch syndrome

Journal of Medical Genetics, Journal Year: 2025, Volume and Issue: unknown, P. jmg - 110656

Published: March 27, 2025

Gene discovery efforts have contributed to a better understanding of the molecular causes dystonia, but knowledge individual monogenic forms remains limited. This review seeks summarise all available data on recently identified autosomal recessive subtype dystonia caused by variants in AOPEP , focusing geographical origins affected families, mutational spectrum, phenotypic expressions and pathophysiology. -related documented as Zech-Boesch syndrome Online Mendelian Inheritance Man database, has been diagnosed cohorts around globe including under-represented populations with increased rates consanguinity. Predictably leading loss protein function, majority (74%) disease-associated alleles are protein-truncating comprising homozygous compound heterozygous stop-gain, frameshift splice-site changes. The dystonic disorder shows onset from childhood fourth decade generalises significant proportion cases (60%). Variable expressivity age-related penetrance likely play role manifestation condition, consistent occasional occurrence pathogenic subjects without diagnosis dystonia. encodes aminopeptidase O, proteolytic processing enzyme that is preferentially expressed glia potentially linked endosomal-lysosomal pathways. worldwide relevance for genetic Future research ˋs cellular metabolism may provide new insights into pathogenesis yet-unidentified therapeutic targets.

Language: Английский

---

Review: Utility of mass spectrometry in rare disease research and diagnosis

npj Genomic Medicine, Journal Year: 2025, Volume and Issue: 10(1)

Published: March 31, 2025

Individuals affected by a rare disease often experience long and arduous diagnostic odyssey. Delivery of genetic answers in timely manner is critical to individuals their families. Multi-omics, term which usually encompasses genomics, transcriptomics, proteomics, metabolomics lipidomics, has gained increasing popularity research diagnosis over the past decade. Mass spectrometry (MS) technique allowing study proteins, metabolites lipids fragments at scale, enabling researchers effectively determine presence abundance thousands molecules single test, accurately quantify specific levels, identify potential therapeutic biomarkers, detect differentially expressed proteins patients with diseases, monitor progression treatment response. In this review, we focus on mass (MS)-based omics survey literature describing utility different MS-based how they have transformed diagnosis.

Language: Английский

---

The prevalence of pathogenic variants in the BMPR2 gene in patients with the idiopathic pulmonary arterial hypertension in the Russian population: sequencing data and meta-analysis

Respiratory Research, Journal Year: 2025, Volume and Issue: 26(1)

Published: April 14, 2025

Idiopathic pulmonary arterial hypertension (IPAH) is a rare and severe form of hypertension, with genetic basis most commonly associated mutations in the BMPR2 gene. However, no testing has been reported for IPAH patients Russian population, nor have systematic studies conducted to assess frequency pathogenic variants this group. The study cohort included 105 patients, consisting 23 males 82 females, who were managed at PH care center Moscow, Russia, from 2014 2024. Genetic was performed using whole-genome sequencing. Variant identification annotation GATK, DeepVariant, VEP, sv-callers AnnotSV. A meta-analysis, MOOSE, 24 involving 3124 470 P/LP variants. Pathogenicity reassessment carried out InterVar, which incorporates ACMG criteria. Analysis adult Russia revealed 11 (10.48%) as carriers or likely pathogenetic (P/LP) As result reassessment, number raised 394 (59%) 445 (67%) 80 became uncertain significance, 152 unclassified P/LP. meta-analysis these reevaluated showed that while our lower than overall average 17.75% difference not statistically significant (p = 0.062). Additionally, we report three variants, literature, one being structural, four TBX4, ATP13A3 AQP1 genes 27 3 patients. For first time, present results population. Despite considerable heterogeneity world-wide data, prevalence population does significantly differ meta-analysis. It crucial periodically reassess pathogenicity published half reclassified LP significance.

Language: Английский

---

Long‐Read Sequencing: The Third Generation of Diagnostic Testing for Dystonia

Movement Disorders, Journal Year: 2025, Volume and Issue: unknown

Published: April 23, 2025

Abstract Long‐read sequencing methodologies provide powerful capacity to identify all types of genomic variations in a single test. platforms such as Oxford Nanopore and PacBio have the potential revolutionize molecular diagnostics by reaching unparalleled accuracies genetic discovery long‐range phasing. In field dystonia, promising results come from recent pilot studies showing improved detection disease‐causing structural variants repeat expansions. Increases throughput ongoing reductions cost will facilitate incorporation long‐read approaches into mainstream diagnostic practice. Although these developments are likely transform clinical care, there is currently discrepancy between benefits application this technique dystonia. review we highlight current opportunities limitations adopting methods for investigation patients with We examples integration evaluation study pathomechanisms individuals dystonic disorders. The goal article stimulate research optimization analysis strategies thus enabling more precise understanding underlying etiology future. © 2025 Author(s). Movement Disorders published Wiley Periodicals LLC on behalf International Parkinson Disorder Society.

Language: Английский

---

Expanding the Allelic and Clinical Heterogeneity of Movement Disorders Linked to Defects of Mitochondrial Adenosine Triphosphate Synthase

Movement Disorders, Journal Year: 2025, Volume and Issue: unknown

Published: April 25, 2025

Abstract Background Defects of mitochondrial ATP synthase (ATPase) represent an emerging, yet incompletely understood group neurodevelopmental diseases with abnormal movements. Objective The aim this study was to redefine the phenotypic and mutational spectrum movement disorders linked ATPase subunit‐encoding genes ATP5F1A ATP5F1B . Methods We recruited regionally distant patients who had been genome or exome sequenced. Fibroblast cultures from two were established perform RNA sequencing, immunoblotting, mass spectrometry–based high‐throughput quantitative proteomics, activity assays. In silico three‐dimensional missense variant modeling performed. Results identified a patient developmental delay, myoclonic dystonia, spasticity carried heterozygous frameshift c.1404del (p.Glu469Serfs*3) in patient's cells exhibited significant reductions mRNA, underexpression α‐subunit association other aberrantly expressed components, compromised activity. addition, novel deleterious c.1252G>A (p.Gly418Arg) discovered, shared by three families hereditary spastic paraplegia (HSP). This mapped functionally important intersubunit communication site. A third variant, c.1074+1G>T, affected canonical donor splice site resulted exon skipping significantly diminished mRNA levels, as well impaired associated phenotype consisted cerebral palsy (CP) prominent generalized dystonia. Conclusions Our data confirm expand role dominant variants disorders. / ‐related should be considered cause HSP, CP. © 2025 Author(s). Movement Disorders published Wiley Periodicals LLC on behalf International Parkinson Disorder Society.

Language: Английский

---

Multimodal AI/ML for discovering novel biomarkers and predicting disease using multi-omics profiles of patients with cardiovascular diseases

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 9, 2024

Abstract Cardiovascular diseases (CVDs) are multifactorial diseases, requiring personalized assessment and treatment. The advancements in multi-omics technologies, namely RNA-seq whole genome sequencing, have offered translational researchers a comprehensive view of the human genome; utilizing this data, we can reveal novel biomarkers segment patient populations based on risk factors. Limitations these technologies failing to capture disease complexity be accounted for by using an integrated approach, characterizing variants alongside expression related emerging phenotypes. Designed implemented data analytics methodology is nexus orthodox bioinformatics, classical statistics, multimodal artificial intelligence machine learning techniques. Our approach has potential intricate mechanisms CVD that facilitate patient-specific response profiling. We sourced transcriptomic from control subjects. By integrating datasets with clinical demographics, generated profiles. Utilizing robust feature selection reported signature 27 transcripts efficient at predicting CVD. Here, differential analysis minimum redundancy maximum relevance elucidated explanatory phenotype. used Combination Annotation Dependent Depletion allele frequencies identify pathogenic characteristics patients. Classification models trained demonstrated high-accuracy predictions CVDs. Overall, observed XGBoost model hyperparameterized Bayesian optimization perform best (AUC 1.0). Using SHapley Additive exPlanations, compiled assessments patients capable further contextualizing setting. discovered 27-component phenotypic differences healthy controls prioritizing both biological efficiency learning. Literature review revealed previous associations majority diagnostic biomarkers. were able predict high accuracy. propose framework generalizable other disorders.

Language: Английский

---

Proteomic Profiling in Dystonia: The Next Frontier for Pathophysiology Research and Biomarker Exploration

Movement Disorders, Journal Year: 2024, Volume and Issue: 39(9), P. 1478 - 1479

Published: Aug. 12, 2024

Data sharing is not applicable to this article as no new data were created or analyzed in study.

Language: Английский

---