A Knowledge-Based Machine Learning Approach to Gene Prioritisation in Amyotrophic Lateral Sclerosis

Identification of proteins' expression pathway and the effective miRNAs for the treatment of human papillomavirus-induced cervical cancer: in-silico analyses-experimental research

Introduction

Cervical cancer is the fourth most common cancer in women. The risk factors for cervical cancer include human papillomavirus (HPV) infection, age, smoking, number of pregnancies, use of oral contraceptives, and diet. However, long-term HPV infection appears to be the main risk factor for developing cervical cancer. This in-silico analysis aims to identify the expression network of proteins and the miRNAs that play a role in the development of HPV-induced cervical cancer.

Methods

The critical proteins and miRNAs were extracted using the DisGeNET and miRBase databases. String and Gephi were applied to the network analysis. The GTEx web tool was utilized to Identify tissue expression levels. The Enrichr website was used to explore the molecular function and pathways of found genes.

Results

Ten proteins, TP53, MYC, AKT1, TNF, IL6, EGFR, STAT3, CTNNB1, ESR1, and JUN, were identified as the most critical shared gene network among cervical cancer and HPV. Seven miRNAs were found, including hsa-mir-146a, hsa-mir-27, hsa-mir-203, hsa-mir-126, hsa-mir-145, hsa-mir-944, and hsa-mir-93, which have a common expression in cervical cancer and HPV.

Conclusion

Overall, the gene network, including TP53, MYC, AKT1, TNF, IL6, EGFR, STAT3, CTNNB1, ESR1, and JUN, and Also, hsa-mir-145, hsa-mir-93, hsa-mir-203, and hsa-mir-126 can be regarded as a gene expression pathway in HPV-induced cervical cancer.

Gene targeting in amyotrophic lateral sclerosis using causality-based feature selection and machine learning

BACKGROUND

Amyotrophic lateral sclerosis (ALS) is a rare progressive neurodegenerative disease that affects upper and lower motor neurons. As the molecular basis of the disease is still elusive, the development of high-throughput sequencing technologies, combined with data mining techniques and machine learning methods, could provide remarkable results in identifying pathogenetic mechanisms. High dimensionality is a major problem when applying machine learning techniques in biomedical data analysis, since a huge number of features is available for a limited number of samples. The aim of this study was to develop a methodology for training interpretable machine learning models in the classification of ALS and ALS-subtypes samples, using gene expression datasets.

METHODS

We performed dimensionality reduction in gene expression data using a semi-automated preprocessing systematic gene selection procedure using Statistically Equivalent Signature (SES), a causality-based feature selection algorithm, followed by Boosted Regression Trees (XGBoost) and Random Forest to train the machine learning classifiers. The SHapley Additive exPlanations (SHAP values) were used for interpretation of the machine learning classifiers. The methodology was developed and tested using two distinct publicly available ALS RNA-seq datasets. We evaluated the performance of SES as a dimensionality reduction method against: (a) Least Absolute Shrinkage and Selection Operator (LASSO), and (b) Local Outlier Factor (LOF).

RESULTS

The proposed methodology achieved 85.18% accuracy for the classification of cerebellum or frontal cortex samples as C9orf72-related familial ALS, sporadic ALS or healthy samples. Importantly, the genes identified as the most determinative have also been reported as disease-associated in ALS literature. When tested in the evaluation dataset, the methodology achieved 88.89% accuracy for the classification of sporadic ALS motor neuron samples. When LASSO was used as feature selection method instead of SES, the accuracy of the machine learning classifiers ranged from 74.07 to 96.30%, depending on tissue assessed, while LOF underperformed significantly (77.78% accuracy for the classification of pooled cerebellum and frontal cortex samples).

CONCLUSIONS

Using SES, we addressed the challenge of high dimensionality in gene expression data analysis, and we trained accurate machine learning ALS classifiers, specific for the gene expression patterns of different disease subtypes and tissue samples, while identifying disease-associated genes.

Single-cell transcriptome reveals core cell populations and androgen-RXFP2 axis involved in deer antler full regeneration

Deer antlers constitute a unique mammalian model for the study of both organ formation in postnatal life and annual full regeneration. Previous studies revealed that these events are achieved through the proliferation and differentiation of antlerogenic periosteum (AP) cells and pedicle periosteum (PP) cells, respectively. As the cells resident in the AP and the PP possess stem cell attributes, both antler generation and regeneration are stem cell-based processes. However, the cell composition of each tissue type and molecular events underlying antler development remain poorly characterized. Here, we took the approach of single-cell RNA sequencing (scRNA-Seq) and identified eight cell types (mainly THY1⁺ cells, progenitor cells, and osteochondroblasts) and three core subclusters of the THY1⁺ cells (SC2, SC3, and SC4). Endothelial and mural cells each are heterogeneous at transcriptional level. It was the proliferation of progenitor, mural, and endothelial cells in the activated antler-lineage-specific tissues that drove the rapid formation of the antler. We detected the differences in the initial differentiation process between antler generation and regeneration using pseudotime trajectory analysis. These may be due to the difference in the degree of stemness of the AP-THY1⁺ and PP-THY1⁺ cells. We further found that androgen-RXFP2 axis may be involved in triggering initial antler full regeneration. Fully deciphering the cell composition for these antler tissue types will open up new avenues for elucidating the mechanism underlying antler full renewal in specific and regenerative medicine in general.

Intersection of network medicine and machine learning towards investigating the key biomarkers and pathways underlying amyotrophic lateral sclerosis: a systematic review

BACKGROUND

Network medicine is an emerging area of research that focuses on delving into the molecular complexity of the disease, leading to the discovery of network biomarkers and therapeutic target discovery. Amyotrophic lateral sclerosis (ALS) is a complicated rare disease with unknown pathogenesis and no available treatment. In ALS, network properties appear to be potential biomarkers that can be beneficial in disease-related applications when explored independently or in tandem with machine learning (ML) techniques.

OBJECTIVE

This systematic literature review explores recent trends in network medicine and implementations of network-based ML algorithms in ALS. We aim to provide an overview of the identified primary studies and gather details on identifying the potential biomarkers and delineated pathways.

METHODS

The current study consists of searching for and investigating primary studies from PubMed and Dimensions.ai, published between 2018 and 2022 that reported network medicine perspectives and the coupling of ML techniques. Each abstract and full-text study was individually evaluated, and the relevant studies were finally included in the review for discussion once they met the inclusion and exclusion criteria.

RESULTS

We identified 109 eligible publications from primary studies representing this systematic review. The data coalesced into two themes: application of network science to identify disease modules and promising biomarkers in ALS, along with network-based ML approaches. Conclusion This systematic review gives an overview of the network medicine approaches and implementations of network-based ML algorithms in ALS to determine new disease genes, and identify critical pathways and therapeutic target discovery for personalized treatment.

Emerging insights into the complex genetics and pathophysiology of amyotrophic lateral sclerosis

Amyotrophic lateral sclerosis is a fatal neurodegenerative disease. The discovery of genes associated with amyotrophic lateral sclerosis, commencing with SOD1 in 1993, started fairly gradually. Recent advances in genetic technology have led to the rapid identification of multiple new genes associated with the disease, and to a new understanding of oligogenic and polygenic disease risk. The overlap of genes associated with amyotrophic lateral sclerosis with those of other neurodegenerative diseases is shedding light on the phenotypic spectrum of neurodegeneration, leading to a better understanding of genotype-phenotype correlations. A deepening knowledge of the genetic architecture is allowing the characterisation of the molecular steps caused by various mutations that converge on recurrent dysregulated pathways. Of crucial relevance, mutations associated with amyotrophic lateral sclerosis are amenable to novel gene-based therapeutic options, an approach in use for other neurological illnesses. Lastly, the exposome-the summation of lifetime environmental exposures-has emerged as an influential component for amyotrophic lateral sclerosis through the gene-time-environment hypothesis. Our improved understanding of all these aspects will lead to long-awaited therapies and the identification of modifiable risks factors.

The panoramic view of amyotrophic lateral sclerosis: A fatal intricate neurological disorder

Amyotrophic lateral sclerosis (ALS) is a progressive and fatal neurological disease affecting both upper and lower motor neurons. In the United States alone, there are 16,000-20,000 established cases of ALS. The early disease diagnosis is challenging due to many overlapping pathophysiologies with other neurological diseases. The etiology of ALS is unknown; however, it is divided into two categories: familial ALS (fALS) which occurs due to gene mutations & contributes to 5-10% of ALS, and sporadic ALS (sALS) which is due to environmental factors & contributes to 90-95% of ALS. There is still no curative treatment for ALS: palliative care and symptomatic treatment are therefore essential components in the management of these patients. In this review, we provide a panoramic view of ALS, which includes epidemiology, risk factors, pathophysiologies, biomarkers, diagnosis, therapeutics (natural, synthetic, gene-based, pharmacological, stem cell, extracellular vesicles, and physical therapy), controversies (in the clinical trials of ALS), the scope of nanomedicine in ALS, and future perspectives.

DGLinker: flexible knowledge-graph prediction of disease-gene associations

As a result of the advent of high-throughput technologies, there has been rapid progress in our understanding of the genetics underlying biological processes. However, despite such advances, the genetic landscape of human diseases has only marginally been disclosed. Exploiting the present availability of large amounts of biological and phenotypic data, we can use our current understanding of disease genetics to train machine learning models to predict novel genetic factors associated with the disease. To this end, we developed DGLinker, a webserver for the prediction of novel candidate genes for human diseases given a set of known disease genes. DGLinker has a user-friendly interface that allows non-expert users to exploit biomedical information from a wide range of biological and phenotypic databases, and/or to upload their own data, to generate a knowledge-graph and use machine learning to predict new disease-associated genes. The webserver includes tools to explore and interpret the results and generates publication-ready figures. DGLinker is available at https://dglinker.rosalind.kcl.ac.uk. The webserver is free and open to all users without the need for registration.

Identification of Five Glycolysis-Related Gene Signature and Risk Score Model for Colorectal Cancer

Metabolic changes, especially in glucose metabolism, are widely established during the occurrence and development of tumors and regarded as biological markers of pan-cancer. The well-known ‘Warburg effect’ demonstrates that cancer cells prefer aerobic glycolysis even if there is sufficient ambient oxygen. Accumulating evidence suggests that aerobic glycolysis plays a pivotal role in colorectal cancer (CRC) development. However, few studies have examined the relationship of glycolytic gene clusters with prognosis of CRC patients. Here, our aim is to build a glycolysis-associated gene signature as a biomarker for colorectal cancer. The mRNA sequencing and corresponding clinical data were downloaded from TCGA and GEO databases. Gene set enrichment analysis (GSEA) was performed, indicating that four gene clusters were significantly enriched, which revealed the inextricable relationship of CRC with glycolysis. By comparing gene expression of cancer and adjacent samples, 236 genes were identified. Univariate, multivariate, and LASSO Cox regression analyses screened out five prognostic-related genes (ENO3, GPC1, P4HA1, SPAG4, and STC2). Kaplan–Meier curves and receiver operating characteristic curves (ROC, AUC = 0.766) showed that the risk model could become an effective prognostic indicator (P < 0.001). Multivariate Cox analysis also revealed that this risk model is independent of age and TNM stages. We further validated this risk model in external cohorts (GES38832 and GSE39582), showing these five glycolytic genes could emerge as reliable predictors for CRC patients’ outcomes. Lastly, based on five genes and risk score, we construct a nomogram model assessed by C-index (0.7905) and calibration plot. In conclusion, we highlighted the clinical significance of glycolysis in CRC and constructed a glycolysis-related prognostic model, providing a promising target for glycolysis regulation in CRC.

Moving Toward Patient-Tailored Treatment in ALS and FTD: The Potential of Genomic Assessment as a Tool for Biological Discovery and Trial Recruitment

Amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) are two devastating and intertwined neurodegenerative diseases. Historically, ALS and FTD were considered distinct disorders given differences in presenting clinical symptoms, disease duration, and predicted risk of developing each disease. However, research over recent years has highlighted the considerable clinical, pathological, and genetic overlap of ALS and FTD, and these two syndromes are now thought to represent different manifestations of the same neuropathological disease spectrum. In this review, we discuss the need to shift our focus from studying ALS and FTD in isolation to identifying the biological mechanisms that drive these diseases-both common and distinct-to improve treatment discovery and therapeutic development success. We also emphasize the importance of genomic data to facilitate a "precision medicine" approach for treating ALS and FTD.

What Can Machine Learning Approaches in Genomics Tell Us about the Molecular Basis of Amyotrophic Lateral Sclerosis?

Amyotrophic Lateral Sclerosis (ALS) is the most common late-onset motor neuron disorder, but our current knowledge of the molecular mechanisms and pathways underlying this disease remain elusive. This review (1) systematically identifies machine learning studies aimed at the understanding of the genetic architecture of ALS, (2) outlines the main challenges faced and compares the different approaches that have been used to confront them, and (3) compares the experimental designs and results produced by those approaches and describes their reproducibility in terms of biological results and the performances of the machine learning models. The majority of the collected studies incorporated prior knowledge of ALS into their feature selection approaches, and trained their machine learning models using genomic data combined with other types of mined knowledge including functional associations, protein-protein interactions, disease/tissue-specific information, epigenetic data, and known ALS phenotype-genotype associations. The importance of incorporating gene-gene interactions and cis-regulatory elements into the experimental design of future ALS machine learning studies is highlighted. Lastly, it is suggested that future advances in the genomic and machine learning fields will bring about a better understanding of ALS genetic architecture, and enable improved personalized approaches to this and other devastating and complex diseases.

From Multi-Omics Approaches to Precision Medicine in Amyotrophic Lateral Sclerosis

Amyotrophic lateral sclerosis (ALS) is a devastating and fatal neurodegenerative disorder, caused by the degeneration of upper and lower motor neurons for which there is no truly effective cure. The lack of successful treatments can be well explained by the complex and heterogeneous nature of ALS, with patients displaying widely distinct clinical features and progression patterns, and distinct molecular mechanisms underlying the phenotypic heterogeneity. Thus, stratifying ALS patients into consistent and clinically relevant subgroups can be of great value for the development of new precision diagnostics and targeted therapeutics for ALS patients. In the last years, the use and integration of high-throughput "omics" approaches have dramatically changed our thinking about ALS, improving our understanding of the complex molecular architecture of ALS, distinguishing distinct patient subtypes and providing a rational foundation for the discovery of biomarkers and new individualized treatments. In this review, we discuss the most significant contributions of omics technologies in unraveling the biological heterogeneity of ALS, highlighting how these approaches are revealing diagnostic, prognostic and therapeutic targets for future personalized interventions.

Genome-wide Meta-analysis Finds the ACSL5-ZDHHC6 Locus Is Associated with ALS and Links Weight Loss to the Disease Genetics

We meta-analyze amyotrophic lateral sclerosis (ALS) genome-wide association study (GWAS) data of European and Chinese populations (84,694 individuals). We find an additional significant association between rs58854276 spanning ACSL5-ZDHHC6 with ALS (p = 8.3 × 10-9), with replication in an independent Australian cohort (1,502 individuals; p = 0.037). Moreover, B4GALNT1, G2E3-SCFD1, and TRIP11-ATXN3 are identified using a gene-based analysis. ACSL5 has been associated with rapid weight loss, as has another ALS-associated gene, GPX3. Weight loss is frequent in ALS patients and is associated with shorter survival. We investigate the effect of the ACSL5 and GPX3 single-nucleotide polymorphisms (SNPs), using longitudinal body composition and weight data of 77 patients and 77 controls. In patients' fat-free mass, although not significant, we observe an effect in the expected direction (rs58854276: -2.1 ± 1.3 kg/A allele, p = 0.053; rs3828599: -1.0 ± 1.3 kg/A allele, p = 0.22). No effect was observed in controls. Our findings support the increasing interest in lipid metabolism in ALS and link the disease genetics to weight loss in patients.