Jingxin Ren

In the context of chronic obstructive pulmonary disease (COPD), understanding gene expression patterns across various cell types is pivotal for disease comprehension, therapeutic development, and enhancement of patient prognosis. Nevertheless, the intricate interplay between immune and structural cells, along with their significance in COPD pathogenesis, remains only partially understood. This research employed multiple machine learning techniques to pinpoint marker genes exhibiting pronounced alterations between COPD patients and healthy individuals. We utilized six feature selection algorithms (LASSO, LightGBM, MCFS, RF, CATboost, and XGBoost) and three classification methods (DT, KNN, and RF) to analyze single-cell patient data. Cells were bifurcated into two subsets: immune cells and epithelial–stromal cells. Each cell, encompassing 21,237 genes, was labeled either as a healthy control or COPD. We identified a number of promising genes, such as FKBP5, which encodes a protein crucial for stress response regulation, and SCGB3A1, which is predominantly expressed in the respiratory tract and critical for regulating lung function. Both genes exhibit strong correlations with COPD. The study also underscored the potential roles of HLA-DRB1 and HSP90AA1 in COPD pathogenesis. Our findings underscore the unparalleled potential of machine learning in identifying genetic markers for intricate diseases like COPD, paving the way for targeted therapeutic advancements.

MicroRNA (miRNA) regulation plays a pivotal role in intracellular gene expression. Analysis of miRNA profiles can provide critical insights into disease states. As cancer-associated molecules reported in previous studies, miRNAs may serve as candidate classificatory features for exploratory cancer classification. This research analyzed serum miRNA data from patients with 13 solid cancer types and individuals without cancer. The study comprised two distinct analyses: first, stratifying the dataset into cancer and non-cancer groups to identify miRNAs differentially represented in cancer patients; and second, subdividing the cancer patient data into 13 predefined solid-cancer types to identify candidate miRNA features that discriminate among these cancer types. We employed seven feature-ranking algorithms to evaluate miRNA contributions in both analyses and generate feature lists. Each list was examined using an incremental feature selection method to extract essential miRNAs and build good-performing classification models. Several candidate miRNAs were identified for distinguishing pan-cancer samples from non-cancer ones: miR-4783-3p has been linked to associated with the regulation of endocrine cell differentiation, and miR-663a has been reported in hepatocellular carcinoma and thyroid carcinoma. The analysis also highlighted miRNAs that differentiate solid cancer types, including miR-629-3p, reported to be upregulated in lung and breast cancer, and miR-6087, reported to be downregulated in osteosarcoma and bladder cancer.

Gastric adenocarcinoma is a significant global health concern. Among the myriad histologic classification methods for this cancer, the Lauren classification stands out as a pivotal tool. Nonetheless, the precision and detail of current histological staging techniques frequently face scrutiny. Utilizing high-resolution single-cell transcriptomic data, this research delves into the distinctive gene expression patterns in diffuse, intestinal, and mixed gastric cancers by deploying various machine learning algorithms. The main goal was to recognize important gene markers and establish efficient classification models. The data was derived from the tumor microenvironment, with cells categorized into six groups according to two locations: tumoral and normal, and three histology types: diffuse, intestinal, and mixed. Every cell describes the expression level of 56,265 genes. We integrated seven feature selection algorithms and four classification algorithms, which increased the accuracy of classification. Importantly, our approach detected intricate expression patterns realized for the first time—for example, high expression of CLDN4 in intestinal-type gastric cancers and CCL4 and CXCR4 in diffuse-type gastric cancers. The identified gene markers and gene expression patterns provide insights into subtype-specific molecular characteristics of gastric cancer. These candidate markers may serve as a foundation for future studies aimed at validating their utility in subtype classification and clinical stratification.

Post-acute sequelae of COVID-19 (PASC) is a persistent complication of severe acute respiratory syndrome coronavirus 2 infection that includes symptoms, such as fatigue, cognitive impairment, and respiratory distress. These symptoms severely affect the quality of life of patients after their recovery from COVID-19. In this study, a group of machine learning algorithms analyzed the whole blood RNA-seq data from patients with different PASC levels. The purpose of this analysis was to identify the gene markers associated with PASC and the special expression patterns for different PASC levels. By comparing the quality of life of patients after the acute phase of COVID-19 and before the disease, samples in the dataset were divided into three groups, namely, “Better,” “The Same,” and “Worse.” Each patient was represented by the expression levels of 58,929 genes. The machine learning-based workflow included six feature-ranking algorithms, incremental feature selection (IFS), and four classification algorithms. The feature ranking algorithms were in charge of assessing feature importance, whereas IFS with classification algorithms were used to extract essential genes and to construct efficient classifiers and classification rules. The expression of top genes in the results was associated with the immune response to viral infection, which is supported by the published literature. For example, patients with low CCDC18 expression and high CPED1 expression had good quality of life, whereas those with low CDC16 expression had poor quality of life. Graphical Abstract

Immune cell infiltration that occurs at the site of colon tumors influences the course of cancer. Different immune cell compositions in the microenvironment lead to different immune responses and different therapeutic effects. This study analyzed single-cell RNA sequencing data in a normal colon with the aim of screening genetic markers of 25 candidate immune cell types and revealing quantitative differences between them. The dataset contains 25 classes of immune cells, 41,650 cells in total, and each cell is expressed by 22,164 genes at the expression level. They were fed into a machine learning-based stream. The five feature ranking algorithms (last absolute shrinkage and selection operator, light gradient boosting machine, Monte Carlo feature selection, minimum redundancy maximum relevance, and random forest) were first used to analyze the importance of gene features, yielding five feature lists. Then, incremental feature selection and two classification algorithms (decision tree and random forest) were combined to filter the most important genetic markers from each list. For different immune cell subtypes, their marker genes, such as KLRB1 in CD4 T cells, RPL30 in B cell IGA plasma cells, and JCHAIN in IgG producing B cells, were identified. They were confirmed to be differentially expressed in different immune cells and involved in immune processes. In addition, quantitative rules were summarized by using the decision tree algorithm to distinguish candidate immune cell types. These results provide a reference for exploring the cell composition of the colon cancer microenvironment and for clinical immunotherapy.

As COVID-19 develops, dynamic changes occur in the patient’s immune system. Changes in molecular levels in different immune cells can reflect the course of COVID-19. This study aims to uncover the molecular characteristics of different immune cell subpopulations at different stages of COVID-19. We designed a machine learning workflow to analyze scRNA-seq data of three immune cell types (B, T, and myeloid cells) in four levels of COVID-19 severity/outcome. The datasets for three cell types included 403,700 B-cell, 634,595 T-cell, and 346,547 myeloid cell samples. Each cell subtype was divided into four groups, control, convalescence, progression mild/moderate, and progression severe/critical, and each immune cell contained 27,943 gene features. A feature analysis procedure was applied to the data of each cell type. Irrelevant features were first excluded according to their relevance to the target variable measured by mutual information. Then, four ranking algorithms (last absolute shrinkage and selection operator, light gradient boosting machine, Monte Carlo feature selection, and max-relevance and min-redundancy) were adopted to analyze the remaining features, resulting in four feature lists. These lists were fed into the incremental feature selection, incorporating three classification algorithms (decision tree, k-nearest neighbor, and random forest) to extract key gene features and construct classifiers with superior performance. The results confirmed that genes such as PFN1, RPS26, and FTH1 played important roles in SARS-CoV-2 infection. These findings provide a useful reference for the understanding of the ongoing effect of COVID-19 development on the immune system.

Corona Virus Disease 2019 (COVID-19) not only causes respiratory system damage, but also imposes strain on the cardiovascular system. Vascular endothelial cells and cardiomyocytes play an important role in cardiac function. The aberrant expression of genes in vascular endothelial cells and cardiomyocytes can lead to cardiovascular diseases. In this study, we sought to explain the influence of respiratory syndrome coronavirus 2 (SARS-CoV-2) infection on the gene expression levels of vascular endothelial cells and cardiomyocytes. We designed an advanced machine learning-based workflow to analyze the gene expression profile data of vascular endothelial cells and cardiomyocytes from patients with COVID-19 and healthy controls. An incremental feature selection method with a decision tree was used in building efficient classifiers and summarizing quantitative classification genes and rules. Some key genes, such as MALAT1, MT-CO1, and CD36, were extracted, which exert important effects on cardiac function, from the gene expression matrix of 104,182 cardiomyocytes, including 12,007 cells from patients with COVID-19 and 92,175 cells from healthy controls, and 22,438 vascular endothelial cells, including 10,812 cells from patients with COVID-19 and 11,626 cells from healthy controls. The findings reported in this study may provide insights into the effect of COVID-19 on cardiac cells and further explain the pathogenesis of COVID-19, and they may facilitate the identification of potential therapeutic targets.

The coronavirus disease 2019 (COVID-19), as a severe respiratory disease, affects many parts of the body, and approximately 20–85% of patients exhibit functional impairment of the senses of smell and taste, some of whom even experience the permanent loss of these senses. These symptoms are not life-threatening but severely affect patients’ quality of life and increase the risk of depression and anxiety. The pathological mechanisms of these symptoms have not been fully identified. In the current study, we aimed to identify the important biomarkers at the expression level associated with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection-mediated loss of taste or olfactory ability, and we have suggested the potential pathogenetic mechanisms of COVID-19 complications. We designed a machine-learning-based approach to analyze the transcriptome of 577 COVID-19 patient samples, including 84 COVID-19 samples with a decreased ability to taste or smell and 493 COVID-19 samples without impairment. Each sample was represented by 58,929 gene expression levels. The features were analyzed and sorted by three feature selection methods (least absolute shrinkage and selection operator, light gradient boosting machine, and Monte Carlo feature selection). The optimal feature sets were obtained through incremental feature selection using two classification algorithms: decision tree (DT) and random forest (RF). The top genes identified by these multiple methods (H3-5, NUDT5, and AOC1) are involved in olfactory and gustatory impairments. Meanwhile, a high-performance RF classifier was developed in this study, and three sets of quantitative rules that describe the impairment of olfactory and gustatory functions were obtained based on the optimal DT classifiers. In summary, this study provides a new computation analysis and suggests the latent biomarkers (genes and rules) for predicting olfactory and gustatory impairment caused by COVID-19 complications.