RGBM: regularized gradient boosting machines for identification of the transcriptional regulators of discrete glioma subtypes

A machine learning approach to differentiate wide QRS tachycardia: distinguishing ventricular tachycardia from supraventricular tachycardia

BACKGROUND

Differential diagnosis of wide QRS tachycardia (WQCT) has been a challenging issue. Published algorithms to distinguish ventricular tachycardia (VT) and supraventricular tachycardia (SVT) have limited diagnostic capabilities.

METHODS

A total of 278 patients with WQCT from January 2010 to March 2022 were enrolled. The electrophysiological study confirmed SVT in 154 patients and VT in 65 ones. Two hundred nineteen WQCT 12-lead ECGs were randomly divided into development cohort (n = 165) and testing cohort (n = 54) data sets. The development cohort was split into a training group (n = 115) and an internal validation group (n = 50). Forty ECG features extracted from the 219 WQCT ECGs are fed into 9 iteratively trained ML algorithms. This novel ML algorithm was also compared with four published algorithms.

RESULTS

In the development cohort, the Gradient Boosting Machine (GBM) model displayed the maximum area under curve (AUC) (0.91, 95% confidence interval (CI) 0.81-1.00). In the testing cohort, the GBM model had a higher AUC of 0.97 compared to 4 validated ECG algorithms, namely, Brugada (0.68), avR (0.62), RWPTII (0.72), and LLA algorithms (0.70). Accuracy, sensitivity, specificity, negative predictive value, and positive predictive value of the GBM model were 0.94, 0.97, 0.90, 0.94, and 0.95, respectively.

CONCLUSIONS

A GBM ML model contributes to distinguishing SVT from VT based on surface ECG features. In addition, we were able to identify important indicators for distinguishing WQCT.

VISH-Pred: an ensemble of fine-tuned ESM models for protein toxicity prediction

Peptide- and protein-based therapeutics are becoming a promising treatment regimen for myriad diseases. Toxicity of proteins is the primary hurdle for protein-based therapies. Thus, there is an urgent need for accurate in silico methods for determining toxic proteins to filter the pool of potential candidates. At the same time, it is imperative to precisely identify non-toxic proteins to expand the possibilities for protein-based biologics. To address this challenge, we proposed an ensemble framework, called VISH-Pred, comprising models built by fine-tuning ESM2 transformer models on a large, experimentally validated, curated dataset of protein and peptide toxicities. The primary steps in the VISH-Pred framework are to efficiently estimate protein toxicities taking just the protein sequence as input, employing an under sampling technique to handle the humongous class-imbalance in the data and learning representations from fine-tuned ESM2 protein language models which are then fed to machine learning techniques such as Lightgbm and XGBoost. The VISH-Pred framework is able to correctly identify both peptides/proteins with potential toxicity and non-toxic proteins, achieving a Matthews correlation coefficient of 0.737, 0.716 and 0.322 and F1-score of 0.759, 0.696 and 0.713 on three non-redundant blind tests, respectively, outperforming other methods by over $10\%$ on these quality metrics. Moreover, VISH-Pred achieved the best accuracy and area under receiver operating curve scores on these independent test sets, highlighting the robustness and generalization capability of the framework. By making VISH-Pred available as an easy-to-use web server, we expect it to serve as a valuable asset for future endeavors aimed at discerning the toxicity of peptides and enabling efficient protein-based therapeutics.

DeepFGRN: inference of gene regulatory network with regulation type based on directed graph embedding

The inference of gene regulatory networks (GRNs) from gene expression profiles has been a key issue in systems biology, prompting many researchers to develop diverse computational methods. However, most of these methods do not reconstruct directed GRNs with regulatory types because of the lack of benchmark datasets or defects in the computational methods. Here, we collect benchmark datasets and propose a deep learning-based model, DeepFGRN, for reconstructing fine gene regulatory networks (FGRNs) with both regulation types and directions. In addition, the GRNs of real species are always large graphs with direction and high sparsity, which impede the advancement of GRN inference. Therefore, DeepFGRN builds a node bidirectional representation module to capture the directed graph embedding representation of the GRN. Specifically, the source and target generators are designed to learn the low-dimensional dense embedding of the source and target neighbors of a gene, respectively. An adversarial learning strategy is applied to iteratively learn the real neighbors of each gene. In addition, because the expression profiles of genes with regulatory associations are correlative, a correlation analysis module is designed. Specifically, this module not only fully extracts gene expression features, but also captures the correlation between regulators and target genes. Experimental results show that DeepFGRN has a competitive capability for both GRN and FGRN inference. Potential biomarkers and therapeutic drugs for breast cancer, liver cancer, lung cancer and coronavirus disease 2019 are identified based on the candidate FGRNs, providing a possible opportunity to advance our knowledge of disease treatments.

CNNGRN: A Convolutional Neural Network-Based Method for Gene Regulatory Network Inference From Bulk Time-Series Expression Data

Gene regulatory networks (GRNs) participate in many biological processes, and reconstructing them plays an important role in systems biology. Although many advanced methods have been proposed for GRN reconstruction, their predictive performance is far from the ideal standard, so it is urgent to design a more effective method to reconstruct GRN. Moreover, most methods only consider the gene expression data, ignoring the network structure information contained in GRN. In this study, we propose a supervised model named CNNGRN, which infers GRN from bulk time-series expression data via convolutional neural network (CNN) model, with a more informative feature. Bulk time series gene expression data imply the intricate regulatory associations between genes, and the network structure feature of ground-truth GRN contains rich neighbor information. Hence, CNNGRN integrates the above two features as model inputs. In addition, CNN is adopted to extract intricate features of genes and infer the potential associations between regulators and target genes. Moreover, feature importance visualization experiments are implemented to seek the key features. Experimental results show that CNNGRN achieved competitive performance on benchmark datasets compared to the state-of-the-art computational methods. Finally, hub genes identified based on CNNGRN have been confirmed to be involved in biological processes through literature.

Pancancer network analysis reveals key master regulators for cancer invasiveness

BACKGROUND

Tumor invasiveness reflects numerous biological changes, including tumorigenesis, progression, and metastasis. To decipher the role of transcriptional regulators (TR) involved in tumor invasiveness, we performed a systematic network-based pan-cancer assessment of master regulators of cancer invasiveness.

MATERIALS AND METHODS

We stratified patients in The Cancer Genome Atlas (TCGA) into invasiveness high (INV-H) and low (INV-L) groups using consensus clustering based on an established robust 24-gene signature to determine the prognostic association of invasiveness with overall survival (OS) across 32 different cancers. We devise a network-based protocol to identify TRs as master regulators (MRs) unique to INV-H and INV-L phenotypes. We validated the activity of MRs coherently associated with INV-H phenotype and worse OS across cancers in TCGA on a series of additional datasets in the Prediction of Clinical Outcomes from the Genomic Profiles (PRECOG) repository.

RESULTS

Based on the 24-gene signature, we defined the invasiveness score for each patient sample and stratified patients into INV-H and INV-L clusters. We observed that invasiveness was associated with worse survival outcomes in almost all cancers and had a significant association with OS in ten out of 32 cancers. Our network-based framework identified common invasiveness-associated MRs specific to INV-H and INV-L groups across the ten prognostic cancers, including COL1A1, which is also part of the 24-gene signature, thus acting as a positive control. Downstream pathway analysis of MRs specific to INV-H phenotype resulted in the identification of several enriched pathways, including Epithelial into Mesenchymal Transition, TGF-β signaling pathway, regulation of Toll-like receptors, cytokines, and inflammatory response, and selective expression of chemokine receptors during T-cell polarization. Most of these pathways have connotations of inflammatory immune response and feasibility for metastasis.

CONCLUSION

Our pan-cancer study provides a comprehensive master regulator analysis of tumor invasiveness and can suggest more precise therapeutic strategies by targeting the identified MRs and downstream enriched pathways for patients across multiple cancers.

Epigenetic and transcriptional activation of the secretory kinase FAM20C as an oncogene in glioma

Gliomas are the most prevalent and aggressive malignancies of the nervous system. Previous bioinformatic studies have revealed the crucial role of the secretory pathway kinase FAM20C in the prediction of glioma invasion and malignancy. However, little is known about the pathogenesis of FAM20C in the regulation of glioma. Here, we construct the full-length transcriptome atlas in paired gliomas and observe that 22 genes are upregulated by full-length transcriptome and differential APA analysis. Analysis of ATAC-seq data reveals that both FAM20C and NPTN are the hub genes with chromatin openness and differential expression. Further, in vitro and in vivo studies suggest that FAM20C stimulates the proliferation and metastasis of glioma cells. Meanwhile, NPTN, a novel cancer suppressor gene, counteracts the function of FAM20C by inhibiting both the proliferation and migration of glioma. The blockade of FAM20C by neutralizing antibodies results in the regression of xenograft tumors. Moreover, MAX, BRD4, MYC, and REST are found to be the potential trans-active factors for the regulation of FAM20C. Taken together, our results uncover the oncogenic role of FAM20C in glioma and shed new light on the treatment of glioma by abolishing FAM20C.

Immunotherapy for brain metastases and primary brain tumors

During the V Siena Immuno-Oncology (IO) Think Tank meeting in 2021, conditions were discussed which favor immunotherapy responses in either primary or secondary brain malignancies. Core elements of these discussions have been reinforced by important publications in 2021 and 2022. In primary brain tumors (such as glioblastoma) current immunotherapies have failed to deliver meaningful clinical benefit. By contrast, brain metastases frequently respond to current immunotherapies. The main differences between both conditions seem to be related to intrinsic factors (e.g., type of driver mutations) and more importantly extrinsic factors, such as the blood brain barrier and immune suppressive microenvironment (e.g., T cell counts, functional differences in T cells, myeloid cells). Future therapeutic interventions may therefore focus on rebalancing the immune cell population in a way which enables the host to respond to current or future immunotherapies.

Pancancer transcriptomic profiling identifies key PANoptosis markers as therapeutic targets for oncology

Resistance to programmed cell death (PCD) is a hallmark of cancer. While some PCD components are prognostic in cancer, the roles of many molecules can be masked by redundancies and crosstalks between PCD pathways, impeding the development of targeted therapeutics. Recent studies characterizing these redundancies have identified PANoptosis, a unique innate immune-mediated inflammatory PCD pathway that integrates components from other PCD pathways. Here, we designed a systematic computational framework to determine the pancancer clinical significance of PANoptosis and identify targetable biomarkers. We found that high expression of PANoptosis genes was detrimental in low grade glioma (LGG) and kidney renal cell carcinoma (KIRC). ZBP1, ADAR, CASP2, CASP3, CASP4, CASP8 and GSDMD expression consistently had negative effects on prognosis in LGG across multiple survival models, while AIM2, CASP3, CASP4 and TNFRSF10 expression had negative effects for KIRC. Conversely, high expression of PANoptosis genes was beneficial in skin cutaneous melanoma (SKCM), with ZBP1, NLRP1, CASP8 and GSDMD expression consistently having positive prognostic effects. As a therapeutic proof-of-concept, we treated melanoma cells with combination therapy that activates ZBP1 and showed that this treatment induced PANoptosis. Overall, through our systematic framework, we identified and validated key innate immune biomarkers from PANoptosis which can be targeted to improve patient outcomes in cancers.

Machine learning for distinguishing right from left premature ventricular contraction origin using surface electrocardiogram features

BACKGROUND

Precise localization of the site of origin of premature ventricular contractions (PVCs) before ablation can facilitate the planning and execution of the electrophysiological procedure.

OBJECTIVE

The purpose of this study was to develop a predictive model that can be used to differentiate PVCs between the left ventricular outflow tract and right ventricular outflow tract (RVOT) using surface electrocardiogram characteristics.

METHODS

A total of 851 patients undergoing radiofrequency ablation of premature ventricular beats from January 2015 to March 2022 were enrolled. Ninety-two patients were excluded. The other 759 patients were enrolled into the development (n = 605), external validation (n = 104), or prospective cohort (n = 50). The development cohort consisted of the training group (n = 423) and the internal validation group (n = 182). Machine learning algorithms were used to construct predictive models for the origin of PVCs using body surface electrocardiogram features.

RESULTS

In the development cohort, the Random Forest model showed a maximum receiver operating characteristic curve area of 0.96. In the external validation cohort, the Random Forest model surpasses 4 reported algorithms in predicting performance (accuracy 94.23%; sensitivity 97.10%; specificity 88.57%). In the prospective cohort, the Random Forest model showed good performance (accuracy 94.00%; sensitivity 85.71%; specificity 97.22%).

CONCLUSION

Random Forest algorithm has improved the accuracy of distinguishing the origin of PVCs, which surpasses 4 previous standards, and would be used to identify the origin of PVCs before the interventional procedure.

Identification and Validation of Immune Markers in Coronary Heart Disease

Background

Coronary heart disease (CHD) is an ischemic heart disease involving a variety of immune factors. This study was aimed at investigating unique immune and m6A patterns in patients with CHD by gene expression in peripheral blood mononuclear cells (PBMCs) and at identifying novel immune biomarkers.

Methods

The CIBERSORT algorithm and single-sample gene set enrichment analysis (ssGSEA) were applied to assess the population of specific infiltrating immunocytes. Weighted Gene Coexpression Network Analysis (WGCNA) was utilized on immune genes matching CHD. A prediction model based on core immune genes was constructed and verified by a machine learning model. Unsupervised cluster analysis identified various immune patterns in the CHD group according to the abundance of immune cells. Methylation of N6 adenosine- (m6A-) related gene was identified from the literature, and t-distributed stochastic neighbor embedding (t-SNE) analysis was used to determine the rationality of the m6A classification. The association between m6A-related genes and various immune cells was estimated using heat maps.

Results

22/28 immune-associated cells differed between the CHD and normal groups, and a significant difference was detected in the expression of 21 m6A-related genes. The proportion of immune-related cells (activated CD4+ T cells and CD8+ T cells) in the peripheral blood of the CHD group was lower than that of the normal group. The immune genes were divided into four modules, of which the turquoise modules showed a significant association with coronary heart disease. Eight hub immune genes (PDGFRA, GNLY, OSMR, NUDT6, FGFR2, IL2RB, TPM2, and S100A1) can well distinguish the CHD group from the normal group. Two different immune patterns were identified in the CHD group. Interestingly, a significant association was detected between the m6A-related genes and immune cell abundance.

Conclusion

In conclusion, we identified different immune and m6A patterns in CHD. Thus, it could be speculated that the immune system plays a crucial role in CHD, and m6A is correlated with immune genes.

Characteristic MicroRNAs Linked to Dysregulated Metabolic Pathways in Qatari Adult Subjects With Obesity and Metabolic Syndrome

BACKGROUND

Obesity-associated dysglycemia is associated with metabolic disorders. MicroRNAs (miRNAs) are known regulators of metabolic homeostasis. We aimed to assess the relationship of circulating miRNAs with clinical features in obese Qatari individuals.

METHODS

We analyzed a dataset of 39 age-matched patients that includes 18 subjects with obesity only (OBO) and 21 subjects with obesity and metabolic syndrome (OBM). We measured 754 well-characterized human microRNAs (miRNAs) and identified differentially expressed miRNAs along with their significant associations with clinical markers in these patients.

RESULTS

A total of 64 miRNAs were differentially expressed between metabolically healthy obese (OBO) versus metabolically unhealthy obese (OBM) patients. Thirteen out of 64 miRNAs significantly correlated with at least one clinical trait of the metabolic syndrome. Six out of the thirteen demonstrated significant association with HbA1c levels; miR-331-3p, miR-452-3p, and miR-485-5p were over-expressed, whereas miR-153-3p, miR-182-5p, and miR-433-3p were under-expressed in the OBM patients with elevated HbA1c levels. We also identified, miR-106b-3p, miR-652-3p, and miR-93-5p that showed a significant association with creatinine; miR-130b-5p, miR-363-3p, and miR-636 were significantly associated with cholesterol, whereas miR-130a-3p was significantly associated with LDL. Additionally, miR-652-3p's differential expression correlated significantly with HDL and creatinine.

CONCLUSIONS

MicroRNAs associated with metabolic syndrome in obese subjects may have a pathophysiologic role and can serve as markers for obese individuals predisposed to various metabolic diseases like diabetes.

Enhancing gene regulatory networks inference through hub-based data integration

One of the main research topics in computational biology is Gene Regulatory Network (GRN) reconstruction that refers to inferring the relationships between genes involved in regulating cell conditions in response to internal or external stimuli. To this end, most computational methods use only transcriptional gene expression data to reconstruct gene regulatory networks, but recent studies suggest that gene expression data must be integrated with other types of data to obtain more accurate models predicting real relationships between genes. In this study, a diffusion-based method is enhanced to integrate biological data of network types besides structural prior knowledge. The Random Walk with Restart algorithm (RWR) with an emphasis on hub nodes is executed separately on each network, and then jointly optimizes low-dimensional feature vectors for network nodes by diffusion component analysis. Next, these feature vectors are used to infer gene regulatory networks. Fourteen centrality measures are studied for the detection of hub nodes to be used in the RWR algorithm, and the best centrality measure having the greatest effect on the improvement of gene network inference is selected. A case study for the Saccharomyces cerevisiae and E. coli networks shows that using the proposed features in comparison with gene expression data alone results in 0.02-0.08 units improvement in Area Under Receiver Characteristic Operator (AUROC) criteria across different gene regulatory network inference methods. Furthermore, the proposed method was applied to the esophageal cancer data to infer its gene regulatory network. The proposed framework substantially improves accuracy and scalability of GRN inference. The fused features and the best centrality measure detected can be used to provide functional insights about genes or proteins in various biological applications. Moreover, it can be served as a general framework for network data and structural data integration and analysis problems in various scientific disciplines including biology.

Simple risk score to screen for prediabetes: A cross-sectional study from the Qatar Biobank cohort

AIMS/INTRODUCTION

The progression from prediabetes to type 2 diabetes is preventable by lifestyle intervention and/or pharmacotherapy in a large fraction of individuals with prediabetes. Our objective was to develop a risk score to screen for prediabetes in the Middle East, where diabetes prevalence is one of the highest in the world.

MATERIALS AND METHODS

In this cross-sectional, case-control study, we used data of 4,895 controls and 2,373 prediabetic adults obtained from the Qatar Biobank cohort. Significant risk factors were identified by logistic regression and other machine learning methods. The receiver operating characteristic was used to calculate the area under curve, cut-off point, sensitivity, specificity, positive and negative predictive values. The prediabetes risk score was developed from data of Qatari citizens, as well as long-term (≥15 years) residents.

RESULTS

The significant risk factors for the Prediabetes Risk Score in Qatar were age, sex, body mass index, waist circumference and blood pressure. The risk score ranges from 0 to 45. The area under the curve of the score was 80% (95% confidence interval 78-83%), and the cut-off point of 16 yielded sensitivity and specificity of 86.2% (95% confidence interval 82.7-89.2%) and 57.9% (95% confidence interval 65.5-71.4%), respectively. Prediabetes Risk Score in Qatar performed equally in Qatari nationals and long-term residents.

CONCLUSIONS

Prediabetes Risk Score in Qatar is the first prediabetes screening score developed in a Middle Eastern population. It only uses risk factors measured non-invasively, is simple, cost-effective, and can be easily understood by the general public and health providers. Prediabetes Risk Score in Qatar is an important tool for early detection of prediabetes, and can help tremendously in curbing the diabetes epidemic in the region.

Network-based identification of key master regulators associated with an immune-silent cancer phenotype

A cancer immune phenotype characterized by an active T-helper 1 (Th1)/cytotoxic response is associated with responsiveness to immunotherapy and favorable prognosis across different tumors. However, in some cancers, such an intratumoral immune activation does not confer protection from progression or relapse. Defining mechanisms associated with immune evasion is imperative to refine stratification algorithms, to guide treatment decisions and to identify candidates for immune-targeted therapy. Molecular alterations governing mechanisms for immune exclusion are still largely unknown. The availability of large genomic datasets offers an opportunity to ascertain key determinants of differential intratumoral immune response. We follow a network-based protocol to identify transcription regulators (TRs) associated with poor immunologic antitumor activity. We use a consensus of four different pipelines consisting of two state-of-the-art gene regulatory network inference techniques, regularized gradient boosting machines and ARACNE to determine TR regulons, and three separate enrichment techniques, including fast gene set enrichment analysis, gene set variation analysis and virtual inference of protein activity by enriched regulon analysis to identify the most important TRs affecting immunologic antitumor activity. These TRs, referred to as master regulators (MRs), are unique to immune-silent and immune-active tumors, respectively. We validated the MRs coherently associated with the immune-silent phenotype across cancers in The Cancer Genome Atlas and a series of additional datasets in the Prediction of Clinical Outcomes from Genomic Profiles repository. A downstream analysis of MRs specific to the immune-silent phenotype resulted in the identification of several enriched candidate pathways, including NOTCH1, TGF-$\beta $, Interleukin-1 and TNF-$\alpha $ signaling pathways. TGFB1I1 emerged as one of the main negative immune modulators preventing the favorable effects of a Th1/cytotoxic response.

A Modelling Framework for Embedding-based Predictions for Compound-Viral Protein Activity

Motivation

A global effort is underway to identify compounds for the treatment of COVID-19. Since de novo compound design is an extremely long, time-consuming, and expensive process, efforts are underway to discover existing compounds that can be repurposed for COVID-19 and new viral diseases.

Model

We propose a machine learning representation framework that uses deep learning induced vector embeddings of compounds and viral proteins as features to predict compound-viral protein activity. The prediction model in-turn uses a consensus framework to rank approved compounds against viral proteins of interest.

Results

Our consensus framework achieves a highmean Pearson correlation of 0.916, mean R2 of 0.840 and a low mean squared error of 0.313 for the task of compound-viral protein activity prediction on an independent test set. As a use case, we identify a ranked list of 47 compounds common to three main proteins of SARS-COV-2 virus (PL-PRO, 3CL-PRO and Spike protein) as potential targets including 21 antivirals, 15 anticancer, 5 antibiotics and 6 other investigationalhuman compounds.We performadditional molecular docking simulations to demonstrate thatmajority of these compounds have low binding energies and thus high binding affinity with the potential to be effective against the SARS-COV-2 virus.

Availability

All the source code and data is available at: https://github.com/raghvendra5688/Drug-Repurposing and https://dx.doi.org/10.17632/8rrwnbcgmx.3. We also implemented a web-server at: https://machinelearning-protein.qcri.org/index.html.

Supplementary information

Supplementary data are available at Bioinformatics online.

Pathway-based classification of glioblastoma uncovers a mitochondrial subtype with therapeutic vulnerabilities

The transcriptomic classification of glioblastoma (GBM) has failed to predict survival and therapeutic vulnerabilities. A computational approach for unbiased identification of core biological traits of single cells and bulk tumors uncovered four tumor cell states and GBM subtypes distributed along neurodevelopmental and metabolic axes, classified as proliferative/progenitor, neuronal, mitochondrial and glycolytic/plurimetabolic. Each subtype was enriched with biologically coherent multiomic features. Mitochondrial GBM was associated with the most favorable clinical outcome. It relied exclusively on oxidative phosphorylation for energy production, whereas the glycolytic/plurimetabolic subtype was sustained by aerobic glycolysis and amino acid and lipid metabolism. Deletion of the glucose-proton symporter SLC45A1 was the truncal alteration most significantly associated with mitochondrial GBM, and the reintroduction of SLC45A1 in mitochondrial glioma cells induced acidification and loss of fitness. Mitochondrial, but not glycolytic/plurimetabolic, GBM exhibited marked vulnerability to inhibitors of oxidative phosphorylation. The pathway-based classification of GBM informs survival and enables precision targeting of cancer metabolism.

Machine Learning to Predict Contrast-Induced Acute Kidney Injury in Patients With Acute Myocardial Infarction

Objective: To develop predictive models for contrast induced acute kidney injury (CI-AKI) among acute myocardial infarction (AMI) patients treated invasively.

Methods: Patients with AMI who underwent angiography therapy were enrolled and randomly divided into training cohort (75%) and validation cohort (25%). Machine learning algorithms were used to construct predictive models for CI-AKI. The predictive models were tested in a validation cohort.

Results: A total of 1,495 patients with AMI were included. Of all the patients, 226 (15.1%) cases developed CI-AKI. In the validation cohort, Random Forest (RF) model with top 15 variables reached an area under the curve (AUC) of 0.82 (95% CI: 0.76–0.87), while the best logistic model had an AUC of 0.69 (95% CI: 0.62–0.76). ACEF (age, creatinine, and ejection fraction) model reached an AUC of 0.62 (95% CI: 0.53–0.71). RF model with top 15 variables achieved a high recall rate of 71.9% and an accuracy of 73.5% in the validation group. Random Forest model significantly outperformed logistic regression in every comparison.

Conclusions: Machine learning algorithms especially Random Forest algorithm improves the accuracy of risk stratifying patients with AMI and should be used to accurately identify the risk of CI-AKI in AMI patients.

Multimodal single-cell analysis reveals distinct radioresistant stem-like and progenitor cell populations in murine glioma

Radiation therapy is part of the standard of care for gliomas and kills a subset of tumor cells, while also altering the tumor microenvironment. Tumor cells with stem-like properties preferentially survive radiation and give rise to glioma recurrence. Various techniques for enriching and quantifying cells with stem-like properties have been used, including the fluorescence activated cell sorting (FACS)-based side population (SP) assay, which is a functional assay that enriches for stem-like tumor cells. In these analyses, mouse models of glioma have been used to understand the biology of this disease and therapeutic responses, including the radiation response. We present combined SP analysis and single-cell RNA sequencing of genetically-engineered mouse models of glioma to show a time course of cellular response to radiation. We identify and characterize two distinct tumor cell populations that are inherently radioresistant and also distinct effects of radiation on immune cell populations within the tumor microenvironment.

Artificial Intelligence (AI) based machine learning models predict glucose variability and hypoglycaemia risk in patients with type 2 diabetes on a multiple drug regimen who fast during ramadan (The PROFAST - IT Ramadan study)

OBJECTIVE

To develop a machine-based algorithm from clinical and demographic data, physical activity and glucose variability to predict hyperglycaemic and hypoglycaemic excursions in patients with type 2 diabetes on multiple glucose lowering therapies who fast during Ramadan.

PATIENTS AND METHODS

Thirteen patients (10 males and three females) with type 2 diabetes on 3 or more anti-diabetic medications were studied with a Fitbit-2 pedometer device and Freestyle Libre (Abbott Diagnostics) 2 weeks before and 2 weeks during Ramadan. Several machine learning techniques were trained to predict blood glucose levels in a regression framework utilising physical activity and contemporaneous blood glucose levels, comparing Ramadan to non-Ramadan days.

RESULTS

The median age of participants was 51 years (IQR 49-52); median BMI was 33.2 kg/m² (IQR 33.0-35.9) and median HbA1c was 7.3% (IQR 6.7-7.8). The optimal model using physical activity achieved an R² of 0.548 and a mean absolute error (MAE) of 30.30. The addition of electronic health record (ehr) information increased R² to 0.636 and reduced MAE to 26.89 and the time of the day feature further increased R² to 0.768 and reduced MAE to 20.55. Combining all the features together resulted in an optimal XGBoost model with an R² of 0.836 and MAE of 17.47. This model accurately estimated normal glucose levels in 2584/2715 (95.2%) readings and hyperglycaemic events in 852/1031 (82.6%) readings, but fewer hypoglycaemic events (48/172 (27.9%)). The optimal XGBoost model prioritized age, gender, BMI and HbA1c followed by glucose levels and physical activity. Interestingly, the blood glucose level prediction by our model was influenced by use of SGLT2i.

CONCLUSION

XGBoost, a machine learning AI algorithm achieves high predictive performance for normal and hyperglycaemic excursions, but has limited predictive value for hypoglycaemia in patients on multiple therapies who fast during Ramadan.

BCrystal: an interpretable sequence-based protein crystallization predictor

MOTIVATION

X-ray crystallography has facilitated the majority of protein structures determined to date. Sequence-based predictors that can accurately estimate protein crystallization propensities would be highly beneficial to overcome the high expenditure, large attrition rate, and to reduce the trial-and-error settings required for crystallization.

RESULTS

In this study, we present a novel model, BCrystal, which uses an optimized gradient boosting machine (XGBoost) on sequence, structural and physio-chemical features extracted from the proteins of interest. BCrystal also provides explanations, highlighting the most important features for the predicted crystallization propensity of an individual protein using the SHAP algorithm. On three independent test sets, BCrystal outperforms state-of-the-art sequence-based methods by more than 12.5% in accuracy, 18% in recall and 0.253 in Matthew's correlation coefficient, with an average accuracy of 93.7%, recall of 96.63% and Matthew's correlation coefficient of 0.868. For relative solvent accessibility of exposed residues, we observed higher values to associate positively with protein crystallizability and the number of disordered regions, fraction of coils and tripeptide stretches that contain multiple histidines associate negatively with crystallizability. The higher accuracy of BCrystal enables it to accurately screen for sequence variants with enhanced crystallizability.

AVAILABILITY AND IMPLEMENTATION

Our BCrystal webserver is at https://machinelearning-protein.qcri.org/ and source code is available at https://github.com/raghvendra5688/BCrystal.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Transcriptional regulatory networks of tumor-associated macrophages that drive malignancy in mesenchymal glioblastoma

BACKGROUND

Glioblastoma (GBM) is a complex disease with extensive molecular and transcriptional heterogeneity. GBM can be subcategorized into four distinct subtypes; tumors that shift towards the mesenchymal phenotype upon recurrence are generally associated with treatment resistance, unfavorable prognosis, and the infiltration of pro-tumorigenic macrophages.

RESULTS

We explore the transcriptional regulatory networks of mesenchymal-associated tumor-associated macrophages (MA-TAMs), which drive the malignant phenotypic state of GBM, and identify macrophage receptor with collagenous structure (MARCO) as the most highly differentially expressed gene. MARCO^high TAMs induce a phenotypic shift towards mesenchymal cellular state of glioma stem cells, promoting both invasive and proliferative activities, as well as therapeutic resistance to irradiation. MARCO^high TAMs also significantly accelerate tumor engraftment and growth in vivo. Moreover, both MA-TAM master regulators and their target genes are significantly correlated with poor clinical outcomes and are often associated with genomic aberrations in neurofibromin 1 (NF1) and phosphoinositide 3-kinases/mammalian target of rapamycin/Akt pathway (PI3K-mTOR-AKT)-related genes. We further demonstrate the origination of MA-TAMs from peripheral blood, as well as their potential association with tumor-induced polarization states and immunosuppressive environments.

CONCLUSIONS

Collectively, our study characterizes the global transcriptional profile of TAMs driving mesenchymal GBM pathogenesis, providing potential therapeutic targets for improving the effectiveness of GBM immunotherapy.

Temporospatial genomic profiling in glioblastoma identifies commonly altered core pathways underlying tumor progression

Background

Tumor heterogeneity underlies resistance and disease progression in glioblastoma (GBM), and tumors most commonly recur adjacent to the surgical resection margins in contrast non-enhancing (NE) regions. To date, no targeted therapies have meaningfully altered overall patient survival in the up-front setting. The aim of this study was to characterize intratumoral heterogeneity in recurrent GBM using bulk samples from primary resection and recurrent samples taken from contrast-enhancing (EN) and contrast NE regions.

Methods

Whole exome and RNA sequencing were performed on matched bulk primary and multiple recurrent EN and NE tumor samples from 16 GBM patients who received standard of care treatment alone or in combination with investigational clinical trial regimens.

Results

Private mutations emerge across multi-region sampling in recurrent tumors. Genomic clonal analysis revealed increased enrichment in gene alterations regulating the G2M checkpoint, Kras signaling, Wnt signaling, and DNA repair in recurrent disease. Subsequent functional studies identified augmented PI3K/AKT transcriptional and protein activity throughout progression, validated by phospho-protein levels. Moreover, a mesenchymal transcriptional signature was observed in recurrent EN regions, which differed from the proneural signature in recurrent NE regions.

Conclusions

Subclonal populations observed within bulk resected primary GBMs transcriptionally evolve across tumor recurrence (EN and NE regions) and exhibit aberrant gene expression of common signaling pathways that persist despite standard or targeted therapy. Our findings provide evidence that there are both adaptive and clonally mediated dependencies of GBM on key pathways, such as the PI3K/AKT axis, for survival across recurrences.

Probing the fibrillation of lysozyme by nanoscale-infrared spectroscopy

Amyloid fibrillation is the root cause of several neuro as well as non-neurological disorders. Understanding the molecular basis of amyloid aggregate formation is crucial for deciphering various neurodegenerative diseases. In our study, we have examined the lysozyme fibrillation process using nano-infrared spectroscopy (nanoIR). NanoIR enabled us to investigate both structural and chemical characteristics of lysozyme fibrillar species concurrently. The spectroscopic results indicate that lysozyme transformed into a fibrillar structure having mainly parallel β-sheets, with almost no antiparallel β-sheets. Features such as protein stiffness have a good correlation with obtained secondary structural information showing the state of the protein within the fibrillation state. The structural and chemical details were compared with transmission electron microscopy (TEM) and circular dichroism (CD). We have utilized nanoIR and measured infrared spectra to characterize lysozyme amyloid fibril structures in terms of morphology, molecular structure, secondary structure content, stability, and size of the cross-β core. We have shown that the use of nanoIR can complement other biophysical studies to analyze the aggregation process and is particularly useful for studying proteins involved in aggregation to help in designing molecules against amyloid aggregation. Specifically, the nanoIR spectra afford higher resolution information and a characteristic fingerprint for determining states of aggregation.Communicated by Ramaswamy H. Sarma.

Cascaded Wx: A Novel Prognosis-Related Feature Selection Framework in Human Lung Adenocarcinoma Transcriptomes

Artificial neural network-based analysis has recently been used to predict clinical outcomes in patients with solid cancers, including lung cancer. However, the majority of algorithms were not originally developed to identify genes associated with patients' prognoses. To address this issue, we developed a novel prognosis-related feature selection framework called Cascaded Wx (CWx). The CWx framework ranks features according to the survival of a given cohort by training neural networks with three different high- and low-risk groups in a cascaded fashion. We showed that this approach accurately identified features that best identify the patients' prognoses, compared to other feature selection algorithms, including the Cox proportional hazards and Coxnet models, when applied to The Cancer Genome Atlas lung adenocarcinoma (LUAD) transcriptome data. The prognostic potential of the top 100 genes identified by CWx outperformed or was comparable to those identified by the other methods as assessed by the concordance index (c-index). In addition, the top 100 genes identified by CWx were found to be associated with the Wnt signaling pathway, providing biologically relevant evidence for the value of these genes in predicting the prognosis of patients with LUAD. Further analyses of other cancer types showed that the genes identified by CWx had the highest prognostic values according to the c-index. Collectively, the CWx framework will potentially be of great use to prognosis-related biomarker discoveries in a variety of diseases.

Benchmark on a large cohort for sleep-wake classification with machine learning techniques

Accurately measuring sleep and its quality with polysomnography (PSG) is an expensive task. Actigraphy, an alternative, has been proven cheap and relatively accurate. However, the largest experiments conducted to date, have had only hundreds of participants. In this work, we processed the data of the recently published Multi-Ethnic Study of Atherosclerosis (MESA) Sleep study to have both PSG and actigraphy data synchronized. We propose the adoption of this publicly available large dataset, which is at least one order of magnitude larger than any other dataset, to systematically compare existing methods for the detection of sleep-wake stages, thus fostering the creation of new algorithms. We also implemented and compared state-of-the-art methods to score sleep-wake stages, which range from the widely used traditional algorithms to recent machine learning approaches. We identified among the traditional algorithms, two approaches that perform better than the algorithm implemented by the actigraphy device used in the MESA Sleep experiments. The performance, in regards to accuracy and F 1 score of the machine learning algorithms, was also superior to the device's native algorithm and comparable to human annotation. Future research in developing new sleep-wake scoring algorithms, in particular, machine learning approaches, will be highly facilitated by the cohort used here. We exemplify this potential by showing that two particular deep-learning architectures, CNN and LSTM, among the many recently created, can achieve accuracy scores significantly higher than other methods for the same tasks.

The molecular landscape of glioma in patients with Neurofibromatosis 1

Neurofibromatosis type 1 (NF1) is a common tumor predisposition syndrome in which glioma is one of the prevalent tumors. Gliomagenesis in NF1 results in a heterogeneous spectrum of low- to high-grade neoplasms occurring during the entire lifespan of patients. The pattern of genetic and epigenetic alterations of glioma that develops in NF1 patients and the similarities with sporadic glioma remain unknown. Here, we present the molecular landscape of low- and high-grade gliomas in patients affected by NF1 (NF1-glioma). We found that the predisposing germline mutation of the NF1 gene was frequently converted to homozygosity and the somatic mutational load of NF1-glioma was influenced by age and grade. High-grade tumors harbored genetic alterations of TP53 and CDKN2A, frequent mutations of ATRX associated with Alternative Lengthening of Telomere, and were enriched in genetic alterations of transcription/chromatin regulation and PI3 kinase pathways. Low-grade tumors exhibited fewer mutations that were over-represented in genes of the MAP kinase pathway. Approximately 50% of low-grade NF1-gliomas displayed an immune signature, T lymphocyte infiltrates, and increased neo-antigen load. DNA methylation assigned NF1-glioma to LGm6, a poorly defined Isocitrate Dehydrogenase 1 wild-type subgroup enriched with ATRX mutations. Thus, the profiling of NF1-glioma defined a distinct landscape that recapitulates a subset of sporadic tumors.

Genexpi: a toolset for identifying regulons and validating gene regulatory networks using time-course expression data

BACKGROUND

Identifying regulons of sigma factors is a vital subtask of gene network inference. Integrating multiple sources of data is essential for correct identification of regulons and complete gene regulatory networks. Time series of expression data measured with microarrays or RNA-seq combined with static binding experiments (e.g., ChIP-seq) or literature mining may be used for inference of sigma factor regulatory networks.

RESULTS

We introduce Genexpi: a tool to identify sigma factors by combining candidates obtained from ChIP experiments or literature mining with time-course gene expression data. While Genexpi can be used to infer other types of regulatory interactions, it was designed and validated on real biological data from bacterial regulons. In this paper, we put primary focus on CyGenexpi: a plugin integrating Genexpi with the Cytoscape software for ease of use. As a part of this effort, a plugin for handling time series data in Cytoscape called CyDataseries has been developed and made available. Genexpi is also available as a standalone command line tool and an R package.

CONCLUSIONS

Genexpi is a useful part of gene network inference toolbox. It provides meaningful information about the composition of regulons and delivers biologically interpretable results.

A metabolic function of FGFR3-TACC3 gene fusions in cancer

Chromosomal translocations that generate in-frame oncogenic gene fusions are notable examples of the success of targeted cancer therapies. We have previously described gene fusions of FGFR3-TACC3 (F3-T3) in 3% of human glioblastoma cases. Subsequent studies have reported similar frequencies of F3-T3 in many other cancers, indicating that F3-T3 is a commonly occuring fusion across all tumour types. F3-T3 fusions are potent oncogenes that confer sensitivity to FGFR inhibitors, but the downstream oncogenic signalling pathways remain unknown. Here we show that human tumours with F3-T3 fusions cluster within transcriptional subgroups that are characterized by the activation of mitochondrial functions. F3-T3 activates oxidative phosphorylation and mitochondrial biogenesis and induces sensitivity to inhibitors of oxidative metabolism. Phosphorylation of the phosphopeptide PIN4 is an intermediate step in the signalling pathway of the activation of mitochondrial metabolism. The F3-T3-PIN4 axis triggers the biogenesis of peroxisomes and the synthesis of new proteins. The anabolic response converges on the PGC1α coactivator through the production of intracellular reactive oxygen species, which enables mitochondrial respiration and tumour growth. These data illustrate the oncogenic circuit engaged by F3-T3 and show that F3-T3-positive tumours rely on mitochondrial respiration, highlighting this pathway as a therapeutic opportunity for the treatment of tumours with F3-T3 fusions. We also provide insights into the genetic alterations that initiate the chain of metabolic responses that drive mitochondrial metabolism in cancer.