A Machine Learning Approach to Parkinson’s Disease Blood Transcriptomics

Artificial intelligence in Parkinson's disease: Early detection and diagnostic advancements

Parkinson's disease (PD) is the second most common neurodegenerative disorder, globally affecting men and women at an exponentially growing rate, with currently no cure. Disease progression starts when dopaminergic neurons begin to die. In PD, the loss of neurotransmitter, dopamine is responsible for the overall communication of neural cells throughout the body. Clinical symptoms of PD are slowness of movement, involuntary muscular contractions, speech & writing changes, lessened automatic movement, and chronic tremors in the body. PD occurs in both familial and sporadic forms and modifiable and non-modifiable risk factors and socioeconomic conditions cause PD. Early detectable diagnostics and treatments have been developed in the last several decades. However, we still do not have precise early detectable biomarkers and therapeutic agents/drugs that prevent and/or delay the disease process. Recently, artificial intelligence (AI) science and machine learning tools have been promising in identifying early detectable markers with a greater rate of accuracy compared to past forms of treatment and diagnostic processes. Artificial intelligence refers to the intelligence exhibited by machines or software, distinct from the intelligence observed in humans that is based on neural networks in a form and can be used to diagnose the longevity and disease severity of disease. The term Machine Learning or Neural Networks is a blanket term used to identify an emerging technology that is created to work in the way of a "human brain" using many intertwined neurons to achieve the same level of raw intelligence as that of a brain. These processes have been used for neurodegenerative diseases such as Parkinson's disease and Alzheimer's disease, to assess the severity of the patient's condition. In the current article, we discuss the prevalence and incidence of PD, and currently available diagnostic biomarkers and therapeutic strategies. We also highlighted currently available artificial intelligence science and machine learning tools and their applications to detect disease and develop therapeutic interventions.

A Systematic Review of Genetics- and Molecular-Pathway-Based Machine Learning Models for Neurological Disorder Diagnosis

The process of identification and management of neurological disorder conditions faces challenges, prompting the investigation of novel methods in order to improve diagnostic accuracy. In this study, we conducted a systematic literature review to identify the significance of genetics- and molecular-pathway-based machine learning (ML) models in treating neurological disorder conditions. According to the study's objectives, search strategies were developed to extract the research studies using digital libraries. We followed rigorous study selection criteria. A total of 24 studies met the inclusion criteria and were included in the review. We classified the studies based on neurological disorders. The included studies highlighted multiple methodologies and exceptional results in treating neurological disorders. The study findings underscore the potential of the existing models, presenting personalized interventions based on the individual's conditions. The findings offer better-performing approaches that handle genetics and molecular data to generate effective outcomes. Moreover, we discuss the future research directions and challenges, emphasizing the demand for generalizing existing models in real-world clinical settings. This study contributes to advancing knowledge in the field of diagnosis and management of neurological disorders.

Alterations in Blood Methylome as Potential Epigenetic Biomarker in Sporadic Parkinson's Disease

OBJECTIVE

To characterize DNA methylation (DNAm) differences between sporadic Parkinson's disease (PD) and healthy control (HC) individuals enrolled in the Parkinson's Progression Markers Initiative (PPMI).

METHODS

Using whole blood, we characterized longitudinal differences in DNAm between sporadic PD patients (n = 196) and HCs (n = 86) enrolled in PPMI. RNA sequencing (RNAseq) was used to conduct gene expression analyses for genes mapped to differentially methylated cytosine-guanine sites (CpGs).

RESULTS

At the time of patient enrollment, 5,178 CpGs were differentially methylated (2,683 hypermethylated and 2,495 hypomethylated) in PD compared to HC. Of these, 579 CpGs underwent significant methylation changes over 3 years. Several differentially methylated CpGs were found near the cytochrome P450 family 2 subfamily E member 1 (CYP2E1) gene. Additionally, multiple hypermethylated CpGs were associated with the N-myc downregulated gene family member 4 (NDRG4) gene. RNA-Seq analyses showed 75 differentially expressed genes in PD patients compared to controls. An integrative analysis of both differentially methylated sites and differentially expressed genes revealed 20 genes that exhibited hypomethylation concomitant with overexpression. Additionally, 1 gene, cathepsin H (CTSH), displayed hypermethylation that was associated with its decreased expression.

INTERPRETATION

We provide initial evidence of alterations in DNAm in blood of PD patients that may serve as potential epigenetic biomarker of disease. To evaluate the significance of these changes throughout the progression of PD, additional profiling at longer intervals and during the prodromal stages of disease will be necessary. ANN NEUROL 2024;95:1162-1172.

A Systematic Review of Extracellular Matrix-Related Alterations in Parkinson's Disease

The role of the extracellular matrix (ECM) in Parkinson's disease (PD) is not well understood, even though it is critical for neuronal structure and signaling. This systematic review identified the top deregulated ECM-related pathways in studies that used gene set enrichment analyses (GSEA) to document transcriptomic, proteomic, or genomic alterations in PD. PubMed and Google scholar were searched for transcriptomics, proteomics, or genomics studies that employed GSEA on data from PD tissues or cells and reported ECM-related pathways among the top-10 most enriched versus controls. Twenty-seven studies were included, two of which used multiple omics analyses. Transcriptomics and proteomics studies were conducted on a variety of tissue and cell types. Of the 17 transcriptomics studies (16 data sets), 13 identified one or more adhesion pathways in the top-10 deregulated gene sets or pathways, primarily related to cell adhesion and focal adhesion. Among the 8 proteomics studies, 5 identified altered overarching ECM gene sets or pathways among the top 10. Among the 4 genomics studies, 3 identified focal adhesion pathways among the top 10. The findings summarized here suggest that ECM organization/structure and cell adhesion (particularly focal adhesion) are altered in PD and should be the focus of future studies.

Federated learning for multi-omics: A performance evaluation in Parkinson's disease

While machine learning (ML) research has recently grown more in popularity, its application in the omics domain is constrained by access to sufficiently large, high-quality datasets needed to train ML models. Federated learning (FL) represents an opportunity to enable collaborative curation of such datasets among participating institutions. We compare the simulated performance of several models trained using FL against classically trained ML models on the task of multi-omics Parkinson's disease prediction. We find that FL model performance tracks centrally trained ML models, where the most performant FL model achieves an AUC-PR of 0.876 ± 0.009, 0.014 ± 0.003 less than its centrally trained variation. We also determine that the dispersion of samples within a federation plays a meaningful role in model performance. Our study implements several open-source FL frameworks and aims to highlight some of the challenges and opportunities when applying these collaborative methods in multi-omics studies.

Construction and analysis of a joint diagnostic model of machine learning for cryptorchidism based on single-cell sequencing

BACKGROUND

Cryptorchidism is a condition in which one or both of a baby's testicles do not fully descend into the bottom of the scrotum. Newborns with cryptorchidism are at increased risk of developing infertility later in life. The aim of this study was to develop a novel diagnostic model for cryptorchidism and to identify new biomarkers associated with cryptorchidism.

METHODS

The study data were obtained from RNA sequencing data of cryptorchid patients from Nantong University Hospital and the Gene Expression Omnibus (GEO) database. Differential expression analysis was used to obtain differentially expressed genes (DEGs) between the control and cryptorchid groups. These DEGs were analyzed for their functions by Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment using GSEA software. Random Forest algorithm was used to screen central genes based on these DEGs. Neuralnet software package was used to develop artificial neural network models. Based on clinical data, receiver operating characteristic (ROC) was used to validate the models. Single-cell sequencing analysis was used for the pathogenesis of cryptorchidism.

RESULTS

We obtained a total of 525 important DEGs related to cryptorchidism, which are mainly associated with biological functions such as supramolecular complexes and microtubule cytoskeleton. Random forest approach screening obtained eight hub genes. A neural network based on the hub genes showed a 100% success rate of the model. Finally, single-cell sequencing analysis validated the hub genes.

CONCLUSION

We developed a novel diagnostic model for cryptorchidism using artificial neural networks and validated its utility as an effective diagnostic tool.

Federated Learning for multi-omics: a performance evaluation in Parkinson's disease

While machine learning (ML) research has recently grown more in popularity, its application in the omics domain is constrained by access to sufficiently large, high-quality datasets needed to train ML models. Federated Learning (FL) represents an opportunity to enable collaborative curation of such datasets among participating institutions. We compare the simulated performance of several models trained using FL against classically trained ML models on the task of multi-omics Parkinson's Disease prediction. We find that FL model performance tracks centrally trained ML models, where the most performant FL model achieves an AUC-PR of 0.876 ± 0.009, 0.014 ± 0.003 less than its centrally trained variation. We also determine that the dispersion of samples within a federation plays a meaningful role in model performance. Our study implements several open source FL frameworks and aims to highlight some of the challenges and opportunities when applying these collaborative methods in multi-omics studies.

Blood transcriptomic signatures associated with molecular changes in the brain and clinical outcomes in Parkinson's disease

The ability to use blood to predict the outcomes of Parkinson's disease, including disease progression and cognitive and motor complications, would be of significant clinical value. We undertook bulk RNA sequencing from the caudate and putamen of postmortem Parkinson's disease (n = 35) and control (n = 40) striatum, and compared molecular profiles with clinical features and bulk RNA sequencing data obtained from antemortem peripheral blood. Cognitive and motor complications of Parkinson's disease were associated with molecular changes in the caudate (stress response) and putamen (endothelial pathways) respectively. Later and earlier-onset Parkinson's disease were molecularly distinct, and disease duration was associated with changes in caudate (oligodendrocyte development) and putamen (cellular senescence), respectively. Transcriptome patterns in the postmortem Parkinson's disease brain were also evident in antemortem peripheral blood, and correlated with clinical features of the disease. Together, these findings identify molecular signatures in Parkinson's disease patients' brain and blood of potential pathophysiologic and prognostic importance.

DNA Methylation Changes in Autologous Hematopoietic Stem Cell Transplant Patients

BACKGROUND

Blood cancers may be potentially cured with hematopoietic stem cell transplantation (HCT); however, standard pre-assessments for transplant eligibility do not capture all contributing factors for transplant outcomes. Epigenetic biomarkers predict outcomes in various diseases. This pilot study aims to explore epigenetic changes (epigenetic age and differentially methylated genes) in patients before and after autologous HCT, that can serve as potential biomarkers to better predict HCT outcomes.

METHODS

This study used a prospective longitudinal study design to compare genome wide DNA methylation changes in 36 autologous HCT eligible patients recruited from the Cellular Immunotherapies and Transplant clinic at a designated National Cancer Center.

RESULTS

Genome-wide DNA methylation, measured by the Illumina Infinium Human Methylation 850K BeadChip, showed a significant difference in DNA methylation patterns post-HCT compared to pre-HCT. Compared to baseline levels of DNA methylation pre-HCT, 3358 CpG sites were hypo-methylated and 3687 were hyper-methylated. Identified differentially methylated positions overlapped with genes involved in hematopoiesis, blood cancers, inflammation and immune responses. Enrichment analyses showed significant alterations in biological processes such as immune response and cell structure organization, however no significant pathways were noted. Though participants had an advanced epigenetic age compared to chronologic age before and after HCT, both epigenetic age and accelerated age decreased post-HCT.

CONCLUSION

Epigenetic changes, both in epigenetic age and differentially methylated genes were observed in autologous HCT recipients, and should be explored as biomarkers to predict transplant outcomes after autologous HCT in larger, longitudinal studies.

Special Issue “Parkinson’s Disease: Genetics and Pathogenesis”

Parkinson’s disease (PD) is a common and incurable neurodegenerative disease, affecting 1% of the population over the age of 65 [...]

Machine Learning Analysis of RNA-seq Data for Diagnostic and Prognostic Prediction of Colon Cancer

Data from omics studies have been used for prediction and classification of various diseases in biomedical and bioinformatics research. In recent years, Machine Learning (ML) algorithms have been used in many different fields related to healthcare systems, especially for disease prediction and classification tasks. Integration of molecular omics data with ML algorithms has offered a great opportunity to evaluate clinical data. RNA sequence (RNA-seq) analysis has been emerged as the gold standard for transcriptomics analysis. Currently, it is being used widely in clinical research. In our present work, RNA-seq data of extracellular vesicles (EV) from healthy and colon cancer patients are analyzed. Our aim is to develop models for prediction and classification of colon cancer stages. Five different canonical ML and Deep Learning (DL) classifiers are used to predict colon cancer of an individual with processed RNA-seq data. The classes of data are formed on the basis of both colon cancer stages and cancer presence (healthy or cancer). The canonical ML classifiers, which are k-Nearest Neighbor (kNN), Logistic Model Tree (LMT), Random Tree (RT), Random Committee (RC), and Random Forest (RF), are tested with both forms of the data. In addition, to compare the performance with canonical ML models, One-Dimensional Convolutional Neural Network (1-D CNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (BiLSTM) DL models are utilized. Hyper-parameter optimizations of DL models are constructed by using genetic meta-heuristic optimization algorithm (GA). The best accuracy in cancer prediction is obtained with RC, LMT, and RF canonical ML algorithms as 97.33%. However, RT and kNN show 95.33% performance. The best accuracy in cancer stage classification is achieved with RF as 97.33%. This result is followed by LMT, RC, kNN, and RT with 96.33%, 96%, 94.66%, and 94%, respectively. According to the results of the experiments with DL algorithms, the best accuracy in cancer prediction is obtained with 1-D CNN as 97.67%. BiLSTM and LSTM show 94.33% and 93.67% performance, respectively. In classification of the cancer stages, the best accuracy is achieved with BiLSTM as 98%. 1-D CNN and LSTM show 97% and 94.33% performance, respectively. The results reveal that both canonical ML and DL models may outperform each other for different numbers of features.

Machine learning within the Parkinson's progression markers initiative: Review of the current state of affairs

The Parkinson's Progression Markers Initiative (PPMI) has collected more than a decade's worth of longitudinal and multi-modal data from patients, healthy controls, and at-risk individuals, including imaging, clinical, cognitive, and 'omics' biospecimens. Such a rich dataset presents unprecedented opportunities for biomarker discovery, patient subtyping, and prognostic prediction, but it also poses challenges that may require the development of novel methodological approaches to solve. In this review, we provide an overview of the application of machine learning methods to analyzing data from the PPMI cohort. We find that there is significant variability in the types of data, models, and validation procedures used across studies, and that much of what makes the PPMI data set unique (multi-modal and longitudinal observations) remains underutilized in most machine learning studies. We review each of these dimensions in detail and provide recommendations for future machine learning work using data from the PPMI cohort.