Feature selection for RNA cleavage efficiency at specific sites using the LASSO regression model in Arabidopsis thaliana

Machine learning-based ensemble prediction model for the gamma passing rate of VMAT-SBRT plan

PURPOSE

The purpose of this study was to accurately predict or classify the beam GPR with an ensemble model by using machine learning for SBRT(VMAT) plans.

METHODS

A total of 128 SBRT VMAT plans with 330 arc beams were retrospectively selected, and 216 radiomics and 34 plan complexity features were calculated for each arc beam. Three models for GPR prediction and classification using support vector machine algorithm were as follows: (1) plan complexity feature-based model (plan model); (2) radiomics feature-based model (radiomics model); and (3) an ensemble model combining the two models (ensemble model). The prediction performance was evaluated by calculating the mean absolute error (MAE), root mean square error (RMSE), and Spearman's correlation coefficient (SC), and the classification performance was measured by calculating the area under the receiver operating characteristic curve (AUC), accuracy, specificity, and sensitivity.

RESULTS

The MAE, RMSE and SC at the 2 %/2 mm gamma criterion in the test dataset were 1.4 %, 2.57 %, and 0.563, respectively, for the plan model; 1.42 %, and 2.51 %, and 0.508, respectively, for the radiomics model; and 1.33 %, 2.49 %, and 0.611, respectively, for the ensemble model. The accuracy, specificity, sensitivity, and AUC at the 2 %/2 mm gamma criterion in the test dataset were 0.807, 0.824, 0.681, and 0.854, respectively, for the plan model; 0.860, 0.893, 0.624, and 0.877, respectively, for the radiomics model; and 0.852, 0.871, 0.710, and 0.896, respectively, for the ensemble model.

CONCLUSIONS

The ensemble model can improve the prediction and classification performance for the GPR of SBRT (VMAT).

Consensus cluster analysis of apoptosis-related genes in patients with osteoarthritis and their correlation with immune cell infiltration

Background

Osteoarthritis (OA) progression involves multiple factors, including cartilage erosion as the basic pathological mechanism of degeneration, and is closely related to chondrocyte apoptosis. To analyze the correlation between apoptosis and OA development, we selected apoptosis genes from the differentially expressed genes (DEGs) between OA and normal samples from the Gene Expression Omnibus (GEO) database, used lasso regression analysis to identify characteristic genes, and performed consensus cluster analysis to further explore the pathogenesis of this disease.

Methods

The Gene expression profile datasets of OA samples, GSE12021 and GSE55235, were downloaded from GEO. The datasets were combined and analyzed for DEGs. Apoptosis-related genes (ARGs) were collected from the GeneCards database and intersected with DEGs for apoptosis-related DEGs (ARDEGs). Least absolute shrinkage and selection operator (LASSO) regression analysis was performed to obtain characteristic genes, and a nomogram was constructed based on these genes. A consensus cluster analysis was performed to divide the patients into clusters. The immune characteristics, functional enrichment, and immune infiltration statuses of the clusters were compared. In addition, a protein-protein interaction network of mRNA drugs, mRNA-transcription factors (TFs), and mRNA-miRNAs was constructed.

Results

A total of 95 DEGs were identified, of which 47 were upregulated and 48 were downregulated, and 31 hub genes were selected as ARDEGs. LASSO regression analysis revealed nine characteristic genes: growth differentiation factor 15 (GDF15), NAMPT, TLR7, CXCL2, KLF2, REV3L, KLF9, THBD, and MTHFD2. Clusters A and B were identified, and neutrophil activation and neutrophil activation involved in the immune response were highly enriched in Cluster B, whereas protein repair and purine salvage signal pathways were enriched in Cluster A. The number of activated natural killer cells in Cluster B was significantly higher than that in Cluster A. GDF15 and KLF9 interacted with 193 and 32 TFs, respectively, and CXCL2 and REV3L interacted with 48 and 82 miRNAs, respectively.

Conclusion

ARGs could predict the occurrence of OA and may be related to different degrees of OA progression.

Model-interpreted outcomes of artificial neural networks classifying immune biomarkers associated with severe infections in ICU

Introduction

Millions of deaths worldwide are a result of sepsis (viral and bacterial) and septic shock syndromes which originate from microbial infections and cause a dysregulated host immune response. These diseases share both clinical and immunological patterns that involve a plethora of biomarkers that can be quantified and used to explain the severity level of the disease. Therefore, we hypothesize that the severity of sepsis and septic shock in patients is a function of the concentration of biomarkers of patients.

Methods

In our work, we quantified data from 30 biomarkers with direct immune function. We used distinct Feature Selection algorithms to isolate biomarkers to be fed into machine learning algorithms, whose mapping of the decision process would allow us to propose an early diagnostic tool.

Results

We isolated two biomarkers, i.e., Programmed Death Ligand-1 and Myeloperoxidase, that were flagged by the interpretation of an Artificial Neural Network. The upregulation of both biomarkers was indicated as contributing to increase the severity level in sepsis (viral and bacterial induced) and septic shock patients.

Discussion

In conclusion, we built a function considering biomarker concentrations to explain severity among sepsis, sepsis COVID, and septic shock patients. The rules of this function include biomarkers with known medical, biological, and immunological activity, favoring the development of an early diagnosis system based in knowledge extracted from artificial intelligence.

Integrated single-cell and bulk RNA sequencing analysis identifies a cancer-associated fibroblast-related gene signature for predicting survival and therapy in gastric cancer

As the dominant component of the tumor microenvironment, cancer-associated fibroblasts (CAFs), play a vital role in tumor progression. An increasing number of studies have confirmed that CAFs are involved in almost every aspect of tumors including tumorigenesis, metabolism, invasion, metastasis and drug resistance, and CAFs provide an attractive therapeutic target. This study aimed to explore the feature genes of CAFs for potential therapeutic targets and reliable prediction of prognosis in patients with gastric cancer (GC). Bioinformatic analysis was utilized to identify the feature genes of CAFs in GC by performing an integrated analysis of single-cell and transcriptome RNA sequencing using R software. Based on these feature genes, a CAF-related gene signature was constructed for prognostic prediction by LASSO. Simultaneously, survival analysis and nomogram were performed to validate the prognostic predictive value of this gene signature, and qRT-PCR and immunohistochemical staining verified the expression of the feature genes of CAFs. In addition, small molecular drugs for gene therapy of CAF-related gene signatures in GC patients were identified using the connectivity map (CMAP) database. A combination of nine CAF-related genes was constructed to characterize the prognosis of GC, and the prognostic potential and differential expression of the gene signature were initially validated. Additionally, three small molecular drugs were deduced to have anticancer properties on GC progression. By integrating single-cell and bulk RNA sequencing analyses, a novel gene signature of CAFs was constructed. The results provide a positive impact on future research and clinical studies involving CAFs for GC.

Combined brain network topological metrics with machine learning algorithms to identify essential tremor

BACKGROUND AND OBJECTIVE

Essential tremor (ET) is a common movement syndrome, and the pathogenesis mechanisms, especially the brain network topological changes in ET are still unclear. The combination of graph theory (GT) analysis with machine learning (ML) algorithms provides a promising way to identify ET from healthy controls (HCs) at the individual level, and further help to reveal the topological pathogenesis in ET.

METHODS

Resting-state functional magnetic resonance imaging (fMRI) data were obtained from 101 ET and 105 HCs. The topological properties were analyzed by using GT analysis, and the topological metrics under every single threshold and the area under the curve (AUC) of all thresholds were used as features. Then a Mann-Whitney U-test and least absolute shrinkage and selection operator (LASSO) were conducted to feature dimensionality reduction. Four ML algorithms were adopted to identify ET from HCs. The mean accuracy, mean balanced accuracy, mean sensitivity, mean specificity, and mean AUC were used to evaluate the classification performance. In addition, correlation analysis was carried out between selected topological features and clinical tremor characteristics.

RESULTS

All classifiers achieved good classification performance. The mean accuracy of Support vector machine (SVM), logistic regression (LR), random forest (RF), and naïve bayes (NB) was 84.65, 85.03, 84.85, and 76.31%, respectively. LR classifier achieved the best classification performance with 85.03% mean accuracy, 83.97% sensitivity, and an AUC of 0.924. Correlation analysis results showed that 2 topological features negatively and 1 positively correlated with tremor severity.

CONCLUSION

These results demonstrated that combining topological metrics with ML algorithms could not only achieve high classification accuracy for discrimination ET from HCs but also help us to reveal the potential topological pathogenesis of ET.

Glycosyltransferase-related long non-coding RNA signature predicts the prognosis of colon adenocarcinoma

Purpose

Colon adenocarcinoma (COAD) is the most common type of colorectal cancer (CRC) and is associated with poor prognosis. Emerging evidence has demonstrated that glycosylation by long noncoding RNAs (lncRNAs) was associated with COAD progression. To date, however, the prognostic values of glycosyltransferase (GT)-related lncRNAs in COAD are still largely unknown.

Methods

We obtained the expression matrix of mRNAs and lncRNAs in COAD from The Cancer Genome Atlas (TCGA) database. Then, the univariate Cox regression analysis was conducted to identify 33 prognostic GT-related lncRNAs. Subsequently, LASSO and multivariate Cox regression analysis were performed, and 7 of 33 GT-related lncRNAs were selected to conduct a risk model. Gene set enrichment analysis (GSEA) was used to analyze gene signaling pathway enrichment of the risk model. ImmuCellAI, an online tool for estimating the abundance of immune cells, and correlation analysis were used to explore the tumor-infiltrating immune cells in COAD. Finally, the expression levels of seven lncRNAs were detected in colorectal cancer cell lines by reverse transcription-quantitative polymerase chain reaction (RT-qPCR).

Results

A total of 1,140 GT-related lncRNAs were identified, and 7 COAD-specific GT-related lncRNAs (LINC02381, MIR210HG, AC009237.14, AC105219.1, ZEB1-AS1, AC002310.1, and AC020558.2) were selected to conduct a risk model. Patients were divided into high- and low-risk groups based on the median of risk score. The prognosis of the high-risk group was worse than that of the low-risk group, indicating the good reliability and specificity of our risk model. Additionally, a nomogram based on the risk score and clinical traits was built to help clinical decisions. GSEA showed that the risk model was significantly enriched in metabolism-related pathways. Immune infiltration analysis revealed that five types of immune cells were significantly different between groups, and two types of immune cells were negatively correlated with the risk score. Besides, we found that the expression levels of these seven lncRNAs in tumor cells were significantly higher than those in normal cells, which verified the feasibility of the risk model.

Conclusion

The efficient risk model based on seven GT-related lncRNAs has prognostic potential for COAD, which may be novel biomarkers and therapeutic targets for COAD patients.

Sequence features around cleavage sites are highly conserved among different species and a critical determinant for RNA cleavage position across eukaryotes

RNA degradation is one of the critical steps for control of gene expression, and endonucleolytic cleavage-dependent RNA degradation is conserved among eukaryotes. Some cleavage sites are secondarily capped in the cytoplasm and identified using the Cap analysis of gene expression (CAGE) method. Although uncapped cleavage sites are widespread in eukaryotes, comparatively little information has been obtained about these sites using CAGE-based degradome analysis. Previously, we developed the truncated RNA-end sequencing (TREseq) method in plant species and used it to acquire comprehensive information about uncapped cleavage sites; we observed G-rich sequences near cleavage sites. However, it remains unclear whether this finding is general to other eukaryotes. In this study, we conducted TREseq analyses in fruit flies (Drosophila melanogaster) and budding yeast (Saccharomyces cerevisiae). The results revealed specific sequence features related to RNA cleavage in D. melanogaster and S. cerevisiae that were similar to sequence patterns in Arabidopsis thaliana. Although previous studies suggest that ribosome movements are important for determining cleavage position, feature selection using a random forest classifier showed that sequences around cleavage sites were major determinant for cleaved or uncleaved sites. Together, our results suggest that sequence features around cleavage sites are critical for determining cleavage position, and that sequence-specific endonucleolytic cleavage-dependent RNA degradation is highly conserved across eukaryotes.

An assessment of meteorological parameters effects on COVID-19 pandemic in Bangladesh using machine learning models

Coronavirus (COVID-19) is a highly contagious virus (SARS-CoV-2) that has caused a global pandemic since January 2020. Scientists around the world are doing extensive research to control this disease. They are working tirelessly to find out the origin and causes of the disease. Several studies and experiments mentioned that there are some meteorological parameters which are highly correlated with COVID-19 transmission. In this work, we studied the effects of 11 meteorological parameters on the transmission of COVID-19 in Bangladesh. We first applied statistical analysis and observed that there is no significant effect of these parameters. Therefore, we proposed a novel technique to analyze the insight effects of these parameters by using a combination of Random Forest, CART, and Lasso feature selection techniques. We observed that 4 parameters are highly influential for COVID-19 where [Formula: see text] and Cloud have positive association whereas WS and AQ have negative impact. Among them, Cloud has the highest positive impact which is 0.063 and WS has the highest negative association which is [Formula: see text]. Moreover, we have validated our performance using DLNM technique. The result of this investigation can be used to develop an alert system that will assist the policymakers to know the characteristics of COVID-19 against meteorological parameters and can impose different policies based on the weather conditions.

RSNET: inferring gene regulatory networks by a redundancy silencing and network enhancement technique

Background

Current gene regulatory network (GRN) inference methods are notorious for a great number of indirect interactions hidden in the predictions. Filtering out the indirect interactions from direct ones remains an important challenge in the reconstruction of GRNs. To address this issue, we developed a redundancy silencing and network enhancement technique (RSNET) for inferring GRNs.

Results

To assess the performance of RSNET method, we implemented the experiments on several gold-standard networks by using simulation study, DREAM challenge dataset and Escherichia coli network. The results show that RSNET method performed better than the compared methods in sensitivity and accuracy. As a case of study, we used RSNET to construct functional GRN for apple fruit ripening from gene expression data.

Conclusions

In the proposed method, the redundant interactions including weak and indirect connections are silenced by recursive optimization adaptively, and the highly dependent nodes are constrained in the model to keep the real interactions. This study provides a useful tool for inferring clean networks.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-022-04696-w.

Methods for detecting RNA degradation intermediates in plants

RNA degradation is an important process for controlling gene expression and is mediated by decapping / deadenylation-dependent or endonucleolytic cleavage-dependent RNA degradation mechanisms. High-throughput sequencing of RNA degradation intermediates was initially developed in Arabidopsis thaliana and similar RNA degradome sequencing methods were conducted in other eukaryotes. However, interpreting results obtained by these sequencing methods is fragmented, and an overview is needed. Here we review the findings and limitations of these sequencing methods and discuss the missing experiments needed to understand RNA degradation intermediates accurately. This review provides direction for future research on RNA degradation and is a reference for RNA degradome studies in other species.

Identification of a Novel Glycosyltransferase Prognostic Signature in Hepatocellular Carcinoma Based on LASSO Algorithm

Although many prognostic models have been developed to help determine personalized prognoses and treatments, the predictive efficiency of these prognostic models in hepatocellular carcinoma (HCC), which is a highly heterogeneous malignancy, is less than ideal. Recently, aberrant glycosylation has been demonstrated to universally participate in tumour initiation and progression, suggesting that dysregulation of glycosyltransferases can serve as novel cancer biomarkers. In this study, a total of 568 RNA-sequencing datasets of HCC from the TCGA database and ICGC database were analysed and integrated via bioinformatic methods. LASSO regression analysis was applied to construct a prognostic signature. Kaplan-Meier survival, ROC curve, nomogram, and univariate and multivariate Cox regression analyses were performed to assess the predictive efficiency of the prognostic signature. GSEA and the "CIBERSORT" R package were utilized to further discover the potential biological mechanism of the prognostic signature. Meanwhile, the differential expression of the prognostic signature was verified by western blot, qRT-PCR and immunohistochemical staining derived from the HPA. Ultimately, we constructed a prognostic signature in HCC based on a combination of six glycosyltransferases, whose prognostic value was evaluated and validated successfully in the testing cohort and the validation cohort. The prognostic signature was identified as an independent unfavourable prognostic factor for OS, and a nomogram including the risk score was established and showed the good performance in predicting OS. Further analysis of the underlying mechanism revealed that the prognostic signature may be potentially associated with metabolic disorders and tumour-infiltrating immune cells.

Machine Learning in the Differentiation of Soft Tissue Neoplasms: Comparison of Fat-Suppressed T2WI and Apparent Diffusion Coefficient (ADC) Features-Based Models

Machine learning has been widely used in the characterization of tumors recently. This article aims to explore the feasibility of the whole tumor fat-suppressed (FS) T2WI and ADC features-based least absolute shrinkage and selection operator (LASSO)-logistic predictive models in the differentiation of soft tissue neoplasms (STN). The clinical and MR findings of 160 cases with 161 histologically proven STN were reviewed, retrospectively, 75 with diffusion-weighted imaging (DWI with b values of 50, 400, and 800 s/mm²). They were divided into benign and malignant groups and further divided into training (70%) and validation (30%) cohorts. The MR FS T2WI and ADC features-based LASSO-logistic models were built and compared. The AUC of the FS T2WI features-based LASSO-logistic regression model for benign and malignant prediction was 0.65 and 0.75 for the training and validation cohorts. The model's sensitivity, specificity, and accuracy of the validation cohort were 55%, 96%, and 76.6%. While the AUC of the ADC features-based model was 0.932 and 0.955 for the training and validation cohorts. The model's sensitivity, specificity, and accuracy were 83.3%, 100%, and 91.7%. The performances of these models were also validated by decision curve analysis (DCA). The AUC of the whole tumor ADC features-based LASSO-logistic regression predictive model was larger than that of FS T2WI features (p = 0.017). The whole tumor fat-suppressed T2WI and ADC features-based LASSO-logistic predictive models both can serve as useful tools in the differentiation of STN. ADC features-based LASSO-logistic regression predictive model did better than that of FS T2WI features.