A random forest based biomarker discovery and power analysis framework for diagnostics research

Tear Proteins Altered in Patients with Persistent Eye Pain after Refractive Surgery: Biomarker Candidate Discovery

Some patients develop persistent eye pain after refractive surgery, but factors that cause or sustain pain are unknown. We tested whether tear proteins of patients with pain 3 months after surgery differ from those of patients without pain. Patients undergoing refractive surgery (laser in situ keratomileusis or photorefractive keratectomy ) were recruited from 2 clinics, and tears were collected 3 months after surgery. Participants rated their eye pain using a numerical rating scale (NRS, 0-10; no pain-worst pain) at baseline, 1 day, and 3 months after surgery. Using tandem mass tag proteomic analysis, we examined tears from patients with pain [NRS ≥ 3 at 3 months (n = 16)] and patients with no pain [NRS ≤ 1 at 3 months (n = 32)] after surgery. A subset of proteins (83 of 2748 detected, 3.0%) were associated with pain 3 months after surgery. High-dimensional statistical models showed that the magnitude of differential expression was not the only important factor in classifying tear samples from pain patients. Models utilizing 3 or 4 proteins had better classification performance than single proteins and represented differences in both directions (higher or lower in pain). Thus, patterns of protein differences may serve as biomarkers of postsurgical eye pain as well as potential therapeutic targets.

Revolutionary Point-of-Care Wearable Diagnostics for Early Disease Detection and Biomarker Discovery through Intelligent Technologies

Early-stage disease detection, particularly in Point-Of-Care (POC) wearable formats, assumes pivotal role in advancing healthcare services and precision-medicine. Public benefits of early detection extend beyond cost-effectively promoting healthcare outcomes, to also include reducing the risk of comorbid diseases. Technological advancements enabling POC biomarker recognition empower discovery of new markers for various health conditions. Integration of POC wearables for biomarker detection with intelligent frameworks represents ground-breaking innovations enabling automation of operations, conducting advanced large-scale data analysis, generating predictive models, and facilitating remote and guided clinical decision-making. These advancements substantially alleviate socioeconomic burdens, creating a paradigm shift in diagnostics, and revolutionizing medical assessments and technology development. This review explores critical topics and recent progress in development of 1) POC systems and wearable solutions for early disease detection and physiological monitoring, as well as 2) discussing current trends in adoption of smart technologies within clinical settings and in developing biological assays, and ultimately 3) exploring utilities of POC systems and smart platforms for biomarker discovery. Additionally, the review explores technology translation from research labs to broader applications. It also addresses associated risks, biases, and challenges of widespread Artificial Intelligence (AI) integration in diagnostics systems, while systematically outlining potential prospects, current challenges, and opportunities.

Enhancing selection of alcohol consumption-associated genes by random forest

Machine learning methods have been used in identifying omics markers for a variety of phenotypes. We aimed to examine whether a supervised machine learning algorithm can improve identification of alcohol-associated transcriptomic markers. In this study, we analysed array-based, whole-blood derived expression data for 17 873 gene transcripts in 5508 Framingham Heart Study participants. By using the Boruta algorithm, a supervised random forest (RF)-based feature selection method, we selected twenty-five alcohol-associated transcripts. In a testing set (30 % of entire study participants), AUC (area under the receiver operating characteristics curve) of these twenty-five transcripts were 0·73, 0·69 and 0·66 for non-drinkers v. moderate drinkers, non-drinkers v. heavy drinkers and moderate drinkers v. heavy drinkers, respectively. The AUC of the selected transcripts by the Boruta method were comparable to those identified using conventional linear regression models, for example, AUC of 1958 transcripts identified by conventional linear regression models (false discovery rate < 0·2) were 0·74, 0·66 and 0·65, respectively. With Bonferroni correction for the twenty-five Boruta method-selected transcripts and three CVD risk factors (i.e. at P < 6·7e-4), we observed thirteen transcripts were associated with obesity, three transcripts with type 2 diabetes and one transcript with hypertension. For example, we observed that alcohol consumption was inversely associated with the expression of DOCK4, IL4R, and SORT1, and DOCK4 and SORT1 were positively associated with obesity, and IL4R was inversely associated with hypertension. In conclusion, using a supervised machine learning method, the RF-based Boruta algorithm, we identified novel alcohol-associated gene transcripts.

Cross-species transcriptomics identifies obesity associated genes between human and mouse studies

BACKGROUND

Fundamentally defined by an imbalance in energy consumption and energy expenditure, obesity is a significant risk factor of several musculoskeletal conditions including osteoarthritis (OA). High-fat diets and sedentary lifestyle leads to increased adiposity resulting in systemic inflammation due to the endocrine properties of adipose tissue producing inflammatory cytokines and adipokines. We previously showed serum levels of specific adipokines are associated with biomarkers of bone remodelling and cartilage volume loss in knee OA patients. Whilst more recently we find the metabolic consequence of obesity drives the enrichment of pro-inflammatory fibroblast subsets within joint synovial tissues in obese individuals compared to those of BMI defined 'health weight'. As such this present study identifies obesity-associated genes in OA joint tissues which are conserved across species and conditions.

METHODS

The study utilised 6 publicly available bulk and single-cell transcriptomic datasets from human and mice studies downloaded from Gene Expression Omnibus (GEO). Machine learning models were employed to model and statistically test datasets for conserved gene expression profiles. Identified genes were validated in OA tissues from obese and healthy weight individuals using quantitative PCR method (N = 38). Obese and healthy-weight patients were categorised by BMI > 30 and BMI between 18 and 24.9 respectively. Informed consent was obtained from all study participants who were scheduled to undergo elective arthroplasty.

RESULTS

Principal component analysis (PCA) was used to investigate the variations between classes of mouse and human data which confirmed variation between obese and healthy populations. Differential gene expression analysis filtered on adjusted p-values of p < 0.05, identified differentially expressed genes (DEGs) in mouse and human datasets. DEGs were analysed further using area under curve (AUC) which identified 12 genes. Pathway enrichment analysis suggests these genes were involved in the biosynthesis and elongation of fatty acids and the transport, oxidation, and catabolic processing of lipids. qPCR validation found the majority of genes showed a tendency to be upregulated in joint tissues from obese participants. Three validated genes, IGFBP2 (p = 0.0363), DOK6 (0.0451) and CASP1 (0.0412) were found to be significantly different in obese joint tissues compared to lean-weight joint tissues.

CONCLUSIONS

The present study has employed machine learning models across several published obesity datasets to identify obesity-associated genes which are validated in joint tissues from OA. These results suggest obesity-associated genes are conserved across conditions and may be fundamental in accelerating disease in obese individuals. Whilst further validations and additional conditions remain to be tested in this model, identifying obesity-associated genes in this way may serve as a global aid for patient stratification giving rise to the potential of targeted therapeutic interventions in such patient subpopulations.

Subtype-WGME enables whole-genome-wide multi-omics cancer subtyping

We present an innovative strategy for integrating whole-genome-wide multi-omics data, which facilitates adaptive amalgamation by leveraging hidden layer features derived from high-dimensional omics data through a multi-task encoder. Empirical evaluations on eight benchmark cancer datasets substantiated that our proposed framework outstripped the comparative algorithms in cancer subtyping, delivering superior subtyping outcomes. Building upon these subtyping results, we establish a robust pipeline for identifying whole-genome-wide biomarkers, unearthing 195 significant biomarkers. Furthermore, we conduct an exhaustive analysis to assess the importance of each omic and non-coding region features at the whole-genome-wide level during cancer subtyping. Our investigation shows that both omics and non-coding region features substantially impact cancer development and survival prognosis. This study emphasizes the potential and practical implications of integrating genome-wide data in cancer research, demonstrating the potency of comprehensive genomic characterization. Additionally, our findings offer insightful perspectives for multi-omics analysis employing deep learning methodologies.

Hybrid mRMR and multi-objective particle swarm feature selection methods and application to metabolomics of traditional Chinese medicine

Metabolomics data has high-dimensional features and a small sample size, which is typical of high-dimensional small sample (HDSS) data. Too high a dimensionality leads to the curse of dimensionality, and too small a sample size tends to trigger overfitting, which poses a challenge to deeper mining in metabolomics. Feature selection is a valuable technique for effectively handling the challenges HDSS data poses. For the feature selection problem of HDSS data in metabolomics, a hybrid Max-Relevance and Min-Redundancy (mRMR) and multi-objective particle swarm feature selection method (MCMOPSO) is proposed. Experimental results using metabolomics data and various University of California, Irvine (UCI) public datasets demonstrate the effectiveness of MCMOPSO in selecting feature subsets with a limited number of high-quality features. MCMOPSO achieves this by efficiently eliminating irrelevant and redundant features, showcasing its efficacy. Therefore, MCMOPSO is a powerful approach for selecting features from high-dimensional metabolomics data with limited sample sizes.

Specific feature recognition on group specific networks (SFR-GSN): a biomarker identification model for cancer stages

Background and Objective

Accurate identification of cancer stages is challenging due to the complexity and heterogeneity of the disease. Current clinical diagnosis methods primarily rely on phenotypic observations, which may not capture early molecular-level changes accurately.

Methods

In this study, a novel biomarker recognition method was proposed tailored for cancer stages by considering the change of gene expression relationships. Utilizing the sample-specific information and protein-protein interaction networks, the group specific networks were constructed to address the limited specificity of potential biomarkers. Then, a specific feature recognition method was proposed based on these group specific networks, which employed the random forest algorithm for initial screening followed by a recursive feature elimination process to identify the optimal biomarker subset. During exploring optimal results, a strategy termed the Cost-Benefit Ratio, was devised to facilitate the identification of stage-specific biomarkers.

Results

Comparative experiments were conducted on lung adenocarcinoma and breast cancer datasets to validate the method's efficacy and generalizability. The results showed that the identified biomarkers were highly stage-specific, and the F1 scores for predicting cancer stages were significantly improved. For the lung adenocarcinoma dataset, the F1 score reached 97.68%, and for the breast cancer dataset, it achieved 96.87%. These results significantly surpassed those of three conventional methods in terms of F1 scores. Moreover, from the perspective of biological functions, the biomarkers were proved playing an important role in cancer stage-evolution.

Conclusion

The proposed method demonstrated its effectiveness in identifying stage-related biomarkers. By using these biomarkers as features, accurate prediction of cancer stages was achieved. Furthermore, the method exhibited potential for biomarker identification in subtype analyses, offering novel perspectives for cancer prognosis.

Evaluation of a proteomic signature coupled with the kidney failure risk equation in predicting end stage kidney disease in a chronic kidney disease cohort

BACKGROUND

The early identification of patients at high-risk for end-stage renal disease (ESRD) is essential for providing optimal care and implementing targeted prevention strategies. While the Kidney Failure Risk Equation (KFRE) offers a more accurate prediction of ESRD risk compared to static eGFR-based thresholds, it does not provide insights into the patient-specific biological mechanisms that drive ESRD. This study focused on evaluating the effectiveness of KFRE in a UK-based advanced chronic kidney disease (CKD) cohort and investigating whether the integration of a proteomic signature could enhance 5-year ESRD prediction.

METHODS

Using the Salford Kidney Study biobank, a UK-based prospective cohort of over 3000 non-dialysis CKD patients, 433 patients met our inclusion criteria: a minimum of four eGFR measurements over a two-year period and a linear eGFR trajectory. Plasma samples were obtained and analysed for novel proteomic signals using SWATH-Mass-Spectrometry. The 4-variable UK-calibrated KFRE was calculated for each patient based on their baseline clinical characteristics. Boruta machine learning algorithm was used for the selection of proteins most contributing to differentiation between patient groups. Logistic regression was employed for estimation of ESRD prediction by (1) proteomic features; (2) KFRE; and (3) proteomic features alongside KFRE.

RESULTS

SWATH maps with 943 quantified proteins were generated and investigated in tandem with available clinical data to identify potential progression biomarkers. We identified a set of proteins (SPTA1, MYL6 and C6) that, when used alongside the 4-variable UK-KFRE, improved the prediction of 5-year risk of ESRD (AUC = 0.75 vs AUC = 0.70). Functional enrichment analysis revealed Rho GTPases and regulation of the actin cytoskeleton pathways to be statistically significant, inferring their role in kidney function and the pathogenesis of renal disease.

CONCLUSIONS

Proteins SPTA1, MYL6 and C6, when used alongside the 4-variable UK-KFRE achieve an improved performance when predicting a 5-year risk of ESRD. Specific pathways implicated in the pathogenesis of podocyte dysfunction were also identified, which could serve as potential therapeutic targets. The findings of our study carry implications for comprehending the involvement of the Rho family GTPases in the pathophysiology of kidney disease, advancing our understanding of the proteomic factors influencing susceptibility to renal damage.

Biomarker discovery with quantum neural networks: a case-study in CTLA4-activation pathways

BACKGROUND

Biomarker discovery is a challenging task due to the massive search space. Quantum computing and quantum Artificial Intelligence (quantum AI) can be used to address the computational problem of biomarker discovery from genetic data.

METHOD

We propose a Quantum Neural Networks architecture to discover genetic biomarkers for input activation pathways. The Maximum Relevance-Minimum Redundancy criteria score biomarker candidate sets. Our proposed model is economical since the neural solution can be delivered on constrained hardware.

RESULTS

We demonstrate the proof of concept on four activation pathways associated with CTLA4, including (1) CTLA4-activation stand-alone, (2) CTLA4-CD8A-CD8B co-activation, (3) CTLA4-CD2 co-activation, and (4) CTLA4-CD2-CD48-CD53-CD58-CD84 co-activation.

CONCLUSION

The model indicates new genetic biomarkers associated with the mutational activation of CLTA4-associated pathways, including 20 genes: CLIC4, CPE, ETS2, FAM107A, GPR116, HYOU1, LCN2, MACF1, MT1G, NAPA, NDUFS5, PAK1, PFN1, PGAP3, PPM1G, PSMD8, RNF213, SLC25A3, UBA1, and WLS. We open source the implementation at: https://github.com/namnguyen0510/Biomarker-Discovery-with-Quantum-Neural-Networks .

Towards identification of postharvest fruit quality transcriptomic markers in Malus domestica

Gene expression is highly impacted by the environment and can be reflective of past events that affected developmental processes. It is therefore expected that gene expression can serve as a signal of a current or future phenotypic traits. In this paper we identify sets of genes, which we call Prognostic Transcriptomic Biomarkers (PTBs), that can predict firmness in Malus domestica (apple) fruits. In apples, all individuals of a cultivar are clones, and differences in fruit quality are due to the environment. The apples transcriptome responds to these differences in environment, which makes PTBs an attractive predictor of future fruit quality. PTBs have the potential to enhance supply chain efficiency, reduce crop loss, and provide higher and more consistent quality for consumers. However, several questions must be addressed. In this paper we answer the question of which of two common modeling approaches, Random Forest or ElasticNet, outperforms the other. We answer if PTBs with few genes are efficient at predicting traits. This is important because we need few genes to perform qPCR, and we answer the question if qPCR is a cost-effective assay as input for PTBs modeled using high-throughput RNA-seq. To do this, we conducted a pilot study using fruit texture in the 'Gala' variety of apples across several postharvest storage regiments. Fruit texture in 'Gala' apples is highly controllable by post-harvest treatments and is therefore a good candidate to explore the use of PTBs. We find that the RandomForest model is more consistent than an ElasticNet model and is predictive of firmness (r2 = 0.78) with as few as 15 genes. We also show that qPCR is reasonably consistent with RNA-seq in a follow up experiment. Results are promising for PTBs, yet more work is needed to ensure that PTBs are robust across various environmental conditions and storage treatments.

The 'analysis of gene expression and biomarkers for point-of-care decision support in Sepsis' study; temporal clinical parameter analysis and validation of early diagnostic biomarker signatures for severe inflammation andsepsis-SIRS discrimination

Introduction

Early diagnosis of sepsis and discrimination from SIRS is crucial for clinicians to provide appropriate care, management and treatment to critically ill patients. We describe identification of mRNA biomarkers from peripheral blood leukocytes, able to identify severe, systemic inflammation (irrespective of origin) and differentiate Sepsis from SIRS, in adult patients within a multi-center clinical study.

Methods

Participants were recruited in Intensive Care Units (ICUs) from multiple UK hospitals, including fifty-nine patients with abdominal sepsis, eighty-four patients with pulmonary sepsis, forty-two SIRS patients with Out-of-Hospital Cardiac Arrest (OOHCA), sampled at four time points, in addition to thirty healthy control donors. Multiple clinical parameters were measured, including SOFA score, with many differences observed between SIRS and sepsis groups. Differential gene expression analyses were performed using microarray hybridization and data analyzed using a combination of parametric and non-parametric statistical tools.

Results

Nineteen high-performance, differentially expressed mRNA biomarkers were identified between control and combined SIRS/Sepsis groups (FC>20.0, p<0.05), termed 'indicators of inflammation' (I°I), including CD177, FAM20A and OLAH. Best-performing minimal signatures e.g. FAM20A/OLAH showed good accuracy for determination of severe, systemic inflammation (AUC>0.99). Twenty entities, termed 'SIRS or Sepsis' (S°S) biomarkers, were differentially expressed between sepsis and SIRS (FC>2·0, p-value<0.05).

Discussion

The best performing signature for discriminating sepsis from SIRS was CMTM5/CETP/PLA2G7/MIA/MPP3 (AUC=0.9758). The I°I and S°S signatures performed variably in other independent gene expression datasets, this may be due to technical variation in the study/assay platform.

Ten metabolites-based algorithm predicts the future development of type 2 diabetes in Chinese

INTRODUCTION

Type 2 diabetes (T2D) is a heterogeneous metabolic disease with large variations in the relative contributions of insulin resistance and β-cell dysfunction across different glucose tolerance subgroups and ethnicities. A more precise yet feasible approach to categorize risk preceding T2D onset is urgently needed. This study aimed to identify potential metabolic biomarkers that could contribute to the development of T2D and investigate whether their impact on T2D is mediated through insulin resistance and β-cell dysfunction.

METHODS

A non-targeted metabolomic analysis was performed in plasma samples of 196 incident T2D cases and 196 age- and sex-matched non-T2D controls recruited from a long-term prospective Chinese community-based cohort with a follow-up period of ∼ 16 years.

RESULTS

Metabolic profiles revealed profound perturbation of metabolomes before T2D onset. Overall metabolic shifts were strongly associated with insulin resistance rather than β-cell dysfunction. In addition, 188 out of the 578 annotated metabolites were associated with insulin resistance. Bi-directional mediation analysis revealed putative causal relationships among the metabolites, insulin resistance and T2D risk. We built a machine-learning based prediction model, integrating the conventional clinical risk factors (age, BMI, TyG index and 2hG) and 10 metabolites (acetyl-tryptophan, kynurenine, γ-glutamyl-phenylalanine, DG(18:2/22:6), DG(38:7), LPI(18:2), LPC(P-16:0), LPC(P-18:1), LPC(P-20:0) and LPE(P-20:0)) (AUROC = 0.894, 5.6% improvement comparing to the conventional clinical risk model), that successfully predicts the development of T2D.

CONCLUSIONS

Our findings support the notion that the metabolic changes resulting from insulin resistance, rather than β-cell dysfunction, are the primary drivers of T2D in Chinese adults. Metabolomes as a valuable phenotype hold potential clinical utility in the prediction of T2D.

Systemic Immune Inflammation Index as a Key Predictor of Dialysis in Pediatric Chronic Kidney Disease with the Use of Random Forest Classifier

BACKGROUND

Low-grade inflammation is a significant component of chronic kidney disease (CKD). Systemic immune inflammation index (SII), a newly defined ratio combining neutrophil, lymphocyte, and platelet counts, has not yet been evaluated in the pediatric CKD population nor in the context of CKD progression or dialysis. Thus, this study aimed to analyze the complete blood cell count (CBC)-driven parameters, including SII, in children with CKD and to assess their potential usefulness in the prediction of the need for chronic dialysis.

METHODS

A single-center, retrospective study was conducted on 27 predialysis children with CKD stages 4-5 and 39 children on chronic dialysis. The data were analyzed with the artificial intelligence tools.

RESULTS

The Random Forest Classifier (RFC) model with the input variables of neutrophil count, mean platelet volume (MPV), and SII turned out to be the best predictor of the progression of pediatric CKD into end-stage kidney disease (ESKD) requiring dialysis. Out of these variables, SII showed the largest share in the prediction of the need for renal replacement therapy.

CONCLUSIONS

Chronic inflammation plays a pivotal role in the progression of CKD into ESKD. Among CBC-driven ratios, SII seems to be the most useful predictor of the need for chronic dialysis in CKD children.

KIM-1, IL-18, and NGAL, in the Machine Learning Prediction of Kidney Injury among Children Undergoing Hematopoietic Stem Cell Transplantation-A Pilot Study

Children undergoing allogeneic hematopoietic stem cell transplantation (HSCT) are prone to developing acute kidney injury (AKI). Markers of kidney damage: kidney injury molecule (KIM)-1, interleukin (IL)-18, and neutrophil gelatinase-associated lipocalin (NGAL) may ease early diagnosis of AKI. The aim of this study was to assess serum concentrations of KIM-1, IL-18, and NGAL in children undergoing HSCT in relation to classical markers of kidney function (creatinine, cystatin C, estimated glomerular filtration rate (eGFR)) and to analyze their usefulness as predictors of kidney damage with the use of artificial intelligence tools. Serum concentrations of KIM-1, IL-18, NGAL, and cystatin C were assessed by ELISA in 27 children undergoing HSCT before transplantation and up to 4 weeks after the procedure. The data was used to build a Random Forest Classifier (RFC) model of renal injury prediction. The RFC model established on the basis of 3 input variables, KIM-1, IL-18, and NGAL concentrations in the serum of children before HSCT, was able to effectively assess the rate of patients with hyperfiltration, a surrogate marker of kidney injury 4 weeks after the procedure. With the use of the RFC model, serum KIM-1, IL-18, and NGAL may serve as markers of incipient renal dysfunction in children after HSCT.

Metabolomic profiling of preterm birth in pregnant women living with HIV

BACKGROUND

Preterm birth is a leading cause of death in children under the age of five. The risk of preterm birth is increased by maternal HIV infection as well as by certain antiretroviral regimens, leading to a disproportionate burden on low- and medium-income settings where HIV is most prevalent. Despite decades of research, the mechanisms underlying spontaneous preterm birth, particularly in resource limited areas with high HIV infection rates, are still poorly understood and accurate prediction and therapeutic intervention remain elusive.

OBJECTIVES

Metabolomics was utilized to identify profiles of preterm birth among pregnant women living with HIV on two different antiretroviral therapy (ART) regimens.

METHODS

This pilot study comprised 100 mother-infant dyads prior to antiretroviral initiation, on zidovudine monotherapy or on protease inhibitor-based antiretroviral therapy. Pregnancies that resulted in preterm births were matched 1:1 with controls by gestational age at time of sample collection. Maternal plasma and blood spots at 23-35 weeks gestation and infant dried blood spots at birth, were assayed using an untargeted metabolomics method. Linear regression and random forests classification models were used to identify shared and treatment-specific markers of preterm birth.

RESULTS

Classification models for preterm birth achieved accuracies of 95.5%, 95.7%, and 80.7% in the untreated, zidovudine monotherapy, and protease inhibitor-based treatment groups, respectively. Urate, methionine sulfone, cortisone, and 17α-hydroxypregnanolone glucuronide were identified as shared markers of preterm birth. Other compounds including hippurate and N-acetyl-1-methylhistidine were found to be significantly altered in a treatment-specific context.

CONCLUSION

This study identified previously known as well as novel metabolomic features of preterm birth in pregnant women living with HIV. Validation of these models in a larger, independent cohort is necessary to ascertain whether they can be utilized to predict preterm birth during a stage of gestation that allows for therapeutic intervention or more effective resource allocation.

Multiomic interpretation of fungus-infected ant metabolomes during manipulated summit disease

Camponotus floridanus ants show altered behaviors followed by a fatal summiting phenotype when infected with manipulating Ophiocordyceps camponoti-floridani fungi. Host summiting as a strategy to increase transmission is also observed with parasite taxa beyond fungi, including aquatic and terrestrial helminths and baculoviruses. The drastic phenotypic changes can sometimes reflect significant molecular changes in gene expression and metabolite concentrations measured in manipulated hosts. Nevertheless, the underlying mechanisms still need to be fully characterized. To investigate the small molecules producing summiting behavior, we infected C. floridanus ants with O. camponoti-floridani and sampled their heads for LC-MS/MS when we observed the characteristic summiting phenotype. We link this metabolomic data with our previous genomic and transcriptomic data to propose mechanisms that underlie manipulated summiting behavior in "zombie ants." This "multiomic" evidence points toward the dysregulation of neurotransmitter levels and neuronal signaling. We propose that these processes are altered during infection and manipulation based on (1) differential expression of neurotransmitter synthesis and receptor genes, (2) altered abundance of metabolites and neurotransmitters (or their precursors) with known behavioral effects in ants and other insects, and (3) possible suppression of a connected immunity pathway. We additionally report signals for metabolic activity during manipulation related to primary metabolism, detoxification, and anti-stress protectants. Taken together, these findings suggest that host manipulation is likely a multi-faceted phenomenon, with key processes changing at multiple levels of molecular organization.

A Machine Learning-Based Analytic Pipeline Applied to Clinical and Serum IgG Immunoproteome Data To Predict Chlamydia trachomatis Genital Tract Ascension and Incident Infection in Women

We developed a reusable and open-source machine learning (ML) pipeline that can provide an analytical framework for rigorous biomarker discovery. We implemented the ML pipeline to determine the predictive potential of clinical and immunoproteome antibody data for outcomes associated with Chlamydia trachomatis (Ct) infection collected from 222 cis-gender females with high Ct exposure. We compared the predictive performance of 4 ML algorithms (naive Bayes, random forest, extreme gradient boosting with linear booster [xgbLinear], and k-nearest neighbors [KNN]), screened from 215 ML methods, in combination with two different feature selection strategies, Boruta and recursive feature elimination. Recursive feature elimination performed better than Boruta in this study. In prediction of Ct ascending infection, naive Bayes yielded a slightly higher median value of are under the receiver operating characteristic curve (AUROC) 0.57 (95% confidence interval [CI], 0.54 to 0.59) than other methods and provided biological interpretability. For prediction of incident infection among women uninfected at enrollment, KNN performed slightly better than other algorithms, with a median AUROC of 0.61 (95% CI, 0.49 to 0.70). In contrast, xgbLinear and random forest had higher predictive performances, with median AUROC of 0.63 (95% CI, 0.58 to 0.67) and 0.62 (95% CI, 0.58 to 0.64), respectively, for women infected at enrollment. Our findings suggest that clinical factors and serum anti-Ct protein IgGs are inadequate biomarkers for ascension or incident Ct infection. Nevertheless, our analysis highlights the utility of a pipeline that searches for biomarkers and evaluates prediction performance and interpretability. IMPORTANCE Biomarker discovery to aid early diagnosis and treatment using machine learning (ML) approaches is a rapidly developing area in host-microbe studies. However, lack of reproducibility and interpretability of ML-driven biomarker analysis hinders selection of robust biomarkers that can be applied in clinical practice. We thus developed a rigorous ML analytical framework and provide recommendations for enhancing reproducibility of biomarkers. We emphasize the importance of robustness in selection of ML methods, evaluation of performance, and interpretability of biomarkers. Our ML pipeline is reusable and open-source and can be used not only to identify host-pathogen interaction biomarkers but also in microbiome studies and ecological and environmental microbiology research.

Identifying potential circulating miRNA biomarkers for the diagnosis and prediction of ovarian cancer using machine-learning approach: application of Boruta

Introduction

In gynecologic oncology, ovarian cancer is a great clinical challenge. Because of the lack of typical symptoms and effective biomarkers for noninvasive screening, most patients develop advanced-stage ovarian cancer by the time of diagnosis. MicroRNAs (miRNAs) are a type of non-coding RNA molecule that has been linked to human cancers. Specifying diagnostic biomarkers to determine non-cancer and cancer samples is difficult.

Methods

By using Boruta, a novel random forest-based feature selection in the machine-learning techniques, we aimed to identify biomarkers associated with ovarian cancer using cancerous and non-cancer samples from the Gene Expression Omnibus (GEO) database: GSE106817. In this study, we used two independent GEO data sets as external validation, including GSE113486 and GSE113740. We utilized five state-of-the-art machine-learning algorithms for classification: logistic regression, random forest, decision trees, artificial neural networks, and XGBoost.

Results

Four models discovered in GSE113486 had an AUC of 100%, three in GSE113740 with AUC of over 94%, and four in GSE113486 with AUC of over 94%. We identified 10 miRNAs to distinguish ovarian cancer cases from normal controls: hsa-miR-1290, hsa-miR-1233-5p, hsa-miR-1914-5p, hsa-miR-1469, hsa-miR-4675, hsa-miR-1228-5p, hsa-miR-3184-5p, hsa-miR-6784-5p, hsa-miR-6800-5p, and hsa-miR-5100. Our findings suggest that miRNAs could be used as possible biomarkers for ovarian cancer screening, for possible intervention.

Hollow Crystallization COF Capsuled MOF Hybrids Depict Serum Metabolic Profiling for Precise Early Diagnosis and Risk Stratification of Acute Coronary Syndrome

Acute coronary syndrome (ACS), comprising unstable angina (UA) and acute myocardial infarction (AMI), is the leading cause of death worldwide. Currently, lacking effective strategies for classifying ACS hinders the prognosis improvement of ACS patients. Disclosing the nature of metabolic disorders holds the potential to reflect disease progress and high-throughput mass spectrometry-based metabolic analysis is a promising tool for large-scale screening. Herein, a hollow crystallization COF capsuled MOF hybrids (UiO-66@HCOF) assisted serum metabolic analysis is developed for the early diagnosis and risk stratification of ACS. UiO-66@HCOF exhibits unrivaled chemical and structural stability as well as endowing satisfying desorption/ionization efficiency in the detection of metabolites. Paired with machine learning algorithms, early diagnosis of ACS is achieved with the area under the curve (AUC) value of 0.945 for validation sets. Besides, a comprehensive ACS risk stratification method is established, and the AUC value for the discrimination of ACS from healthy controls, and AMI from UA are 0.890, and 0.928. Moreover, the AUC value of the subtyping of AMI is 0.964. Finally, the potential biomarkers exhibit high sensitivity and specificity. This study makes metabolic molecular diagnosis a reality and provided new insight into the progress of ACS.

Exploring the Synergistic Potential of Radiomics and Laboratory Biomarkers for Enhanced Identification of Vulnerable COVID-19 Patients

BACKGROUND

Severe courses and high hospitalization rates were ubiquitous during the first pandemic SARS-CoV-2 waves. Thus, we aimed to examine whether integrative diagnostics may aid in identifying vulnerable patients using crucial data and materials obtained from COVID-19 patients hospitalized between 2020 and 2021 (n = 52). Accordingly, we investigated the potential of laboratory biomarkers, specifically the dynamic cell decay marker cell-free DNA and radiomics features extracted from chest CT.

METHODS

Separate forward and backward feature selection was conducted for linear regression with the Intensive-Care-Unit (ICU) period as the initial target. Three-fold cross-validation was performed, and collinear parameters were reduced. The model was adapted to a logistic regression approach and verified in a validation naïve subset to avoid overfitting.

RESULTS

The adapted integrated model classifying patients into "ICU/no ICU demand" comprises six radiomics and seven laboratory biomarkers. The models' accuracy was 0.54 for radiomics, 0.47 for cfDNA, 0.74 for routine laboratory, and 0.87 for the combined model with an AUC of 0.91.

CONCLUSION

The combined model performed superior to the individual models. Thus, integrating radiomics and laboratory data shows synergistic potential to aid clinic decision-making in COVID-19 patients. Under the need for evaluation in larger cohorts, including patients with other SARS-CoV-2 variants, the identified parameters might contribute to the triage of COVID-19 patients.

Proteomic signature associated with chronic kidney disease (CKD) progression identified by data-independent acquisition mass spectrometry

BACKGROUND

Halting progression of chronic kidney disease (CKD) to established end stage kidney disease is a major goal of global health research. The mechanism of CKD progression involves pro-inflammatory, pro-fibrotic, and vascular pathways, but pathophysiological differentiation is currently lacking.

METHODS

Plasma samples of 414 non-dialysis CKD patients, 170 fast progressors (with ∂ eGFR-3 ml/min/1.73 m2/year or worse) and 244 stable patients (∂ eGFR of - 0.5 to + 1 ml/min/1.73 m2/year) with a broad range of kidney disease aetiologies, were obtained and interrogated for proteomic signals with SWATH-MS. We applied a machine learning approach to feature selection of proteins quantifiable in at least 20% of the samples, using the Boruta algorithm. Biological pathways enriched by these proteins were identified using ClueGo pathway analyses.

RESULTS

The resulting digitised proteomic maps inclusive of 626 proteins were investigated in tandem with available clinical data to identify biomarkers of progression. The machine learning model using Boruta Feature Selection identified 25 biomarkers as being important to progression type classification (Area Under the Curve = 0.81, Accuracy = 0.72). Our functional enrichment analysis revealed associations with the complement cascade pathway, which is relevant to CKD as the kidney is particularly vulnerable to complement overactivation. This provides further evidence to target complement inhibition as a potential approach to modulating the progression of diabetic nephropathy. Proteins involved in the ubiquitin-proteasome pathway, a crucial protein degradation system, were also found to be significantly enriched.

CONCLUSIONS

The in-depth proteomic characterisation of this large-scale CKD cohort is a step toward generating mechanism-based hypotheses that might lend themselves to future drug targeting. Candidate biomarkers will be validated in samples from selected patients in other large non-dialysis CKD cohorts using a targeted mass spectrometric analysis.

Neoepitope load, T cell signatures and PD-L2 as combined biomarker strategy for response to checkpoint inhibition immunotherapy

Immune checkpoint inhibition for the treatment of cancer has provided a breakthrough in oncology, and several new checkpoint inhibition pathways are currently being investigated regarding their potential to provide additional clinical benefit. However, only a fraction of patients respond to such treatment modalities, and there is an urgent need to identify biomarkers to rationally select patients that will benefit from treatment. In this study, we explore different tumor associated characteristics for their association with favorable clinical outcome in a diverse cohort of cancer patients treated with checkpoint inhibitors. We studied 29 patients in a basket trial comprising 12 different tumor types, treated with 10 different checkpoint inhibition regimens. Our analysis revealed that even across this diverse cohort, patients achieving clinical benefit had significantly higher neoepitope load, higher expression of T cell signatures, and higher PD-L2 expression, which also correlated with improved progression-free and overall survival. Importantly, the combination of biomarkers serves as a better predictor than each of the biomarkers alone. Basket trials are frequently used in modern immunotherapy trial design, and here we identify a set of biomarkers of potential relevance across multiple cancer types, allowing for the selection of patients that most likely will benefit from immune checkpoint inhibition.

Humoral SARS-CoV-2 Immune Response in COVID-19 Recovered Vaccinated and Unvaccinated Individuals Related to Post-COVID-Syndrome

BACKGROUND

The duration of anti-SARS-CoV-2-antibody detectability up to 12 months was examined in individuals after either single convalescence or convalescence and vaccination. Moreover, variables that might influence an anti-RBD/S1 antibody decline and the existence of a post-COVID-syndrome (PCS) were addressed.

METHODS

Forty-nine SARS-CoV-2-qRT-PCR-confirmed participants completed a 12-month examination of anti-SARS-CoV-2-antibody levels and PCS-associated long-term sequelae. Overall, 324 samples were collected. Cell-free DNA (cfDNA) was isolated and quantified from EDTA-plasma. As cfDNA is released into the bloodstream from dying cells, it might provide information on organ damage in the late recovery of COIVD-19. Therefore, we evaluated cfDNA concentrations as a biomarker for a PCS. In the context of antibody dynamics, a random forest-based logistic regression with antibody decline as the target was performed and internally validated.

RESULTS

The mean percentage dynamic related to the maximum measured value was 96 (±38)% for anti-RBD/S1 antibodies and 30 (±26)% for anti-N antibodies. Anti-RBD/S1 antibodies decreased in 37%, whereas anti-SARS-CoV-2-anti-N antibodies decreased in 86% of the subjects. Clinical anti-RBD/S1 antibody decline prediction models, including vascular and other diseases, were cross-validated (highest AUC 0.74). Long-term follow-up revealed no significant reduction in PCS prevalence but an increase in cognitive impairment, with no indication for cfDNA as a marker for a PCS.

CONCLUSION

Long-term anti-RBD/S1-antibody positivity was confirmed, and clinical parameters associated with declining titers were presented. A fulminant decrease in anti-SARS-CoV-2-anti-N antibodies was observed (mean change to maximum value 30 (±26)%). Anti-RBD/S1 antibody titers of SARS-CoV-2 recovered subjects boosted with a vaccine exceeded the maximum values measured after single infection by 235 ± 382-fold, with no influence on preexisting PCS. PCS long-term prevalence was 38.6%, with an increase in cognitive impairment compromising the quality of life. Quantified cfDNA measured in the early post-COVID-19 phase might not be an effective marker for PCS identification.

Associating transcriptomics data with inflammatory markers to understand tumour microenvironment in hepatocellular carcinoma

BACKGROUND

Liver cancer is the fourth leading cause of cancer-related death globally which is estimated to reach more than 1 million deaths a year by 2030. Among liver cancer types, hepatocellular carcinoma (HCC) accounts for approximately 90% of the cases and is known to have a tumour promoting inflammation regardless of its underlying aetiology. However, current promising treatment approaches, such as immunotherapy, are partially effective for most of the patients due to the immunosuppressive nature of the tumour microenvironment (TME). Therefore, there is an urgent need to fully understand TME in HCC and discover new immune markers to eliminate resistance to immunotherapy.

METHODS

We analyse three microarray datasets, using unsupervised and supervised methods, in an effort to discover signature genes. First, univariate, and multivariate, feature selection methods, such as the Boruta algorithm, are applied. Subsequently, an optimisation procedure, which utilises random forest algorithm with three dataset pairs combinations, is performed. The resulting optimal gene sets are then combined and further subjected to network analysis and pathway enrichment analysis so as to obtain information related to their biological relevance. The microarray datasets were analysed via the MCP-counter, CIBERSORT, TIMER, EPIC, and quanTIseq deconvolution methods and an estimation of cell type abundances for each dataset sample were identified. The differences in the cell type abundances, between the adjacent and tumour sample groups, were then assessed using a Wilcoxon Rank Sum test (p-value < 0.05).

RESULTS

The optimal gene signature sets, derived from each of the data pairs combination, achieved AUC values ranging from 0.959 to 0.988 in external validation sets using Random Forest model. CLEC1B and PTTG1 genes are retrieved across each optimal set. Among the signature genes, PTTG1, AURKA, and UBE2C genes are found to be involved in the regulation of mitotic sister chromatid separation and anaphase-promoting complex (APC) dependent catabolic process (adjusted p-value < 0.001). Additionally, the application of deconvolution algorithms revealed significant changes in cell type abundances of Regulatory T (Treg) cells, M0 and M1 macrophages, and T CD8+ cells between adjacent and tumour samples.

CONCLUSION

We identified ECM1 gene as a potential immune-related marker acting through immune cell migration and macrophage polarisation. Our results indicate that macrophages, such as M0 macrophage and M1 macrophage cells, undergo significant changes in HCC TME. Moreover, our immune deconvolution approach revealed significant infiltration of Treg cells and M0 macrophages, and a significant decrease in T CD8+ cells and M1 macrophages in tumour samples.

Polymorphisms predicting phylogeny in hepatitis B virus

Hepatitis B viruses (HBVs) are compact viruses with circular genomes of ∼3.2 kb in length. Four genes (HBx, Core, Surface, and Polymerase) generating seven products are encoded on overlapping reading frames. Ten HBV genotypes have been characterised (A-J), which may account for differences in transmission, outcomes of infection, and treatment response. However, HBV genotyping is rarely undertaken, and sequencing remains inaccessible in many settings. We set out to assess which amino acid (aa) sites in the HBV genome are most informative for determining genotype, using a machine learning approach based on random forest algorithms (RFA). We downloaded 5,496 genome-length HBV sequences from a public database, excluding recombinant sequences, regions with conserved indels, and genotypes I and J. Each gene was separately translated into aa, and the proteins concatenated into a single sequence (length 1,614 aa). Using RFA, we searched for aa sites predictive of genotype and assessed covariation among the sites with a mutual information-based method. We were able to discriminate confidently between genotypes A-H using ten aa sites. Half of these sites (5/10) sites were identified in Polymerase (Pol), of which 4/5 were in the spacer domain and one in reverse transcriptase. A further 4/10 sites were located in Surface protein and a single site in HBx. There were no informative sites in Core. Properties of the aa were generally not conserved between genotypes at informative sites. Among the highest co-varying pairs of sites, there were fifty-five pairs that included one of these 'top ten' sites. Overall, we have shown that RFA analysis is a powerful tool for identifying aa sites that predict the HBV lineage, with an unexpectedly high number of such sites in the spacer domain, which has conventionally been viewed as unimportant for structure or function. Our results improve ease of genotype prediction from limited regions of HBV sequences and may have future applications in understanding HBV evolution.

Specific triacylglycerol, diacylglycerol, and lyso-phosphatidylcholine species for the prediction of type 2 diabetes: a ~ 16-year prospective study in Chinese

Background

Bioactive lipids play an important role in insulin secretion and sensitivity, contributing to the pathophysiology of type 2 diabetes (T2D). This study aimed to identify novel lipid species associated with incident T2D in a nested case–control study within a long-term prospective Chinese community-based cohort with a median follow-up of ~ 16 years.

Methods

Plasma samples from 196 incident T2D cases and 196 age- and sex-matched non-T2D controls recruited from the Hong Kong Cardiovascular Risk Factor Prevalence Study (CRISPS) were first analyzed using untargeted lipidomics. Potential predictive lipid species selected by the Boruta analysis were then verified by targeted lipidomics. The associations between these lipid species and incident T2D were assessed. Effects of novel lipid species on insulin secretion in mouse islets were investigated.

Results

Boruta analysis identified 16 potential lipid species. After adjustment for body mass index (BMI), triacylglycerol/high-density lipoprotein (TG/HDL) ratio and the presence of prediabetes, triacylglycerol (TG) 12:0_18:2_22:6, TG 16:0_11:1_18:2, TG 49:0, TG 51:1 and diacylglycerol (DG) 18:2_22:6 were independently associated with increased T2D risk, whereas lyso-phosphatidylcholine (LPC) O-16:0, LPC P-16:0, LPC O-18:0 and LPC 18:1 were independently associated with decreased T2D risk. Addition of the identified lipid species to the clinical prediction model, comprised of BMI, TG/HDL ratio and the presence of prediabetes, achieved a 3.8% improvement in the area under the receiver operating characteristics curve (AUROC) (p = 0.0026). Further functional study revealed that, LPC O-16:0 and LPC O-18:0 significantly potentiated glucose induced insulin secretion (GSIS) in a dose-dependent manner, whereas neither DG 18:2_22:6 nor TG 12:0_18:2_22:6 had any effect on GSIS.

Conclusions

Addition of the lipid species substantially improved the prediction of T2D beyond the model based on clinical risk factors. Decreased levels of LPC O-16:0 and LPC O-18:0 may contribute to the development of T2D via reduced insulin secretion.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12933-022-01677-4.

Investigation of the Role of PUFA Metabolism in Breast Cancer Using a Rank-Based Random Forest Algorithm

Simple Summary

Polyunsaturated fatty acids (PUFAs) and their derivatives, oxylipins, are a constant focus of cancer research due to the relationship between cancer and processes of energy metabolism and inflammation, where a PUFA system is an active player. Only recently have methods been developed that allow for studying such complex systems. Using the Rank-based Random Forest (RF) model, we show that PUFA metabolism genes are critical for the pathogenesis of breast cancer (BC); BC subtypes differ in PUFA metabolism gene expression. The enrichment of BC subtypes with various genes associated with oxylipin signaling pathways indicates a different contribution of these compounds to the biology of subtypes.

Abstract

Polyunsaturated fatty acid (PUFA) metabolism is currently a focus in cancer research due to PUFAs functioning as structural components of the membrane matrix, as fuel sources for energy production, and as sources of secondary messengers, so called oxylipins, important players of inflammatory processes. Although breast cancer (BC) is the leading cause of cancer death among women worldwide, no systematic study of PUFA metabolism as a system of interrelated processes in this disease has been carried out. Here, we implemented a Boruta-based feature selection algorithm to determine the list of most important PUFA metabolism genes altered in breast cancer tissues compared with in normal tissues. A rank-based Random Forest (RF) model was built on the selected gene list (33 genes) and applied to predict the cancer phenotype to ascertain the PUFA genes involved in cancerogenesis. It showed high-performance of dichotomic classification (balanced accuracy of 0.94, ROC AUC 0.99) We also retrieved a list of the important PUFA genes (46 genes) that differed between molecular subtypes at the level of breast cancer molecular subtypes. The balanced accuracy of the classification model built on the specified genes was 0.82, while the ROC AUC for the sensitivity analysis was 0.85. Specific patterns of PUFA metabolic changes were obtained for each molecular subtype of breast cancer. These results show evidence that (1) PUFA metabolism genes are critical for the pathogenesis of breast cancer; (2) BC subtypes differ in PUFA metabolism genes expression; and (3) the lists of genes selected in the models are enriched with genes involved in the metabolism of signaling lipids.

C10Pred: A First Machine Learning Based Tool to Predict C10 Family Cysteine Peptidases Using Sequence-Derived Features

Streptococcus pyogenes, or group A Streptococcus (GAS), a gram-positive bacterium, is implicated in a wide range of clinical manifestations and life-threatening diseases. One of the key virulence factors of GAS is streptopain, a C10 family cysteine peptidase. Since its discovery, various homologs of streptopain have been reported from other bacterial species. With the increased affordability of sequencing, a significant increase in the number of potential C10 family-like sequences in the public databases is anticipated, posing a challenge in classifying such sequences. Sequence-similarity-based tools are the methods of choice to identify such streptopain-like sequences. However, these methods depend on some level of sequence similarity between the existing C10 family and the target sequences. Therefore, in this work, we propose a novel predictor, C10Pred, for the prediction of C10 peptidases using sequence-derived optimal features. C10Pred is a support vector machine (SVM) based model which is efficient in predicting C10 enzymes with an overall accuracy of 92.7% and Matthews’ correlation coefficient (MCC) value of 0.855 when tested on an independent dataset. We anticipate that C10Pred will serve as a handy tool to classify novel streptopain-like proteins belonging to the C10 family and offer essential information.

Untargeted serum metabolomic profiling for early detection of Schistosoma mekongi infection in mouse model

Mekong schistosomiasis is a parasitic disease caused by blood flukes in the Lao People’s Democratic Republic and in Cambodia. The standard method for diagnosis of schistosomiasis is detection of parasite eggs from patient samples. However, this method is not sufficient to detect asymptomatic patients, low egg numbers, or early infection. Therefore, diagnostic methods with higher sensitivity at the early stage of the disease are needed to fill this gap. The aim of this study was to identify potential biomarkers of early schistosomiasis using an untargeted metabolomics approach. Serum of uninfected and S. mekongi-infected mice was collected at 2, 4, and 8 weeks post-infection. Samples were extracted for metabolites and analyzed with a liquid chromatography-tandem mass spectrometer. Metabolites were annotated with the MS-DIAL platform and analyzed with Metaboanalyst bioinformatic tools. Multivariate analysis distinguished between metabolites from the different experimental conditions. Biomarker screening was performed using three methods: correlation coefficient analysis; feature important detection with a random forest algorithm; and receiver operating characteristic (ROC) curve analysis. Three compounds were identified as potential biomarkers at the early stage of the disease: heptadecanoyl ethanolamide; picrotin; and theophylline. The levels of these three compounds changed significantly during early-stage infection, and therefore these molecules may be promising schistosomiasis markers. These findings may help to improve early diagnosis of schistosomiasis, thus reducing the burden on patients and limiting spread of the disease in endemic areas.

Forecasting Staphylococcus aureus Infections Using Genome-Wide Association Studies, Machine Learning, and Transcriptomic Approaches

Staphylococcus aureus is a major human and animal pathogen, colonizing diverse ecological niches within its hosts. Predicting whether an isolate will infect a specific host and its subsequent clinical fate remains unknown. In this study, we investigated the S. aureus pangenome using a curated set of 356 strains, spanning a wide range of hosts, origins, and clinical display and antibiotic resistance profiles. We used genome-wide association study (GWAS) and random forest (RF) algorithms to discriminate strains based on their origins and clinical sources. Here, we show that the presence of sak and scn can discriminate strains based on their host specificity, while other genes such as mecA are often associated with virulent outcomes. Both GWAS and RF indicated the importance of intergenic regions (IGRs) and coding DNA sequence (CDS) but not sRNAs in forecasting an outcome. Additional transcriptomic analyses performed on the most prevalent clonal complex 8 (CC8) clonal types, in media mimicking nasal colonization or bacteremia, indicated three RNAs as potential RNA markers to forecast infection, followed by 30 others that could serve as infection severity predictors. Our report shows that genetic association and transcriptomics are complementary approaches that will be combined in a single analytical framework to improve our understanding of bacterial pathogenesis and ultimately identify potential predictive molecular markers.

IMPORTANCE Predicting the outcome of bacterial colonization and infections, based on extensive genomic and transcriptomic data from a given pathogen, would be of substantial help for clinicians in treating and curing patients. In this report, genome-wide association studies and random forest algorithms have defined gene combinations that differentiate human from animal strains, colonization from diseases, and nonsevere from severe diseases, while it revealed the importance of IGRs and CDS, but not small RNAs (sRNAs), in anticipating an outcome. In addition, transcriptomic analyses performed on the most prevalent clonal types, in media mimicking either nasal colonization or bacteremia, revealed significant differences and therefore potent RNA markers. Overall, the use of both genomic and transcriptomic data in a single analytical framework can enhance our understanding of bacterial pathogenesis.

Anosmia Testing as Early Detection of SARS-CoV-2 Positivity; A Prospective Study under Screening Conditions

Sudden onset of anosmia is a phenomenon often associated with developing COVID-19 disease and has even been described as an initial isolated symptom in individual cases. In this case-control study, we investigated the feasibility of this condition as a suitable screening test in a population at risk. We performed a prospective study with a total of 313 subjects with suspected SARS-CoV-2 infection. In parallel to routine PCR analysis, a modified commercial scent test was performed to objectify the presence of potential anosmia as a predictor of SARS-CoV-2 positivity. Furthermore, a structured interview assessment of the participants was conducted. A total of 12.1% of the study participants had molecular genetic detection of SARS-CoV-2 infection in the nasopharyngeal swab. It could be demonstrated that these subjects had a significantly weaker olfactory identification performance of the scents. Further analysis of the collected data from the scent test and medical history via random forest (Boruta) algorithm showed that no improvement of the prediction power was achieved by this design. The assay investigated in this study may be suitable for screening general olfactory function. For the screening of COVID-19, it seems to be affected by too many external and internal biases and requires too elaborate and selective pre-test screening.

Vascular Dysfunction Is Central to Alzheimer's Disease Pathogenesis in APOE e4 Carriers

Alzheimer's disease (AD) is the most common form of dementia and the leading risk factor, after age, is possession of the apolipoprotein E epsilon 4 allele (APOE4). Approximately 50% of AD patients carry one or two copies of APOE4 but the mechanisms by which it confers risk are still unknown. APOE4 carriers are reported to demonstrate changes in brain structure, cognition, and neuropathology, but findings have been inconsistent across studies. In the present study, we used multi-modal data to characterise the effects of APOE4 on the brain, to investigate whether AD pathology manifests differently in APOE4 carriers, and to determine if AD pathomechanisms are different between carriers and non-carriers. Brain structural differences in APOE4 carriers were characterised by applying machine learning to over 2000 brain MRI measurements from 33,384 non-demented UK biobank study participants. APOE4 carriers showed brain changes consistent with vascular dysfunction, such as reduced white matter integrity in posterior brain regions. The relationship between APOE4 and AD pathology was explored among the 1260 individuals from the Religious Orders Study and Memory and Aging Project (ROSMAP). APOE4 status had a greater effect on amyloid than tau load, particularly amyloid in the posterior cortical regions. APOE status was also highly correlated with cerebral amyloid angiopathy (CAA). Bulk tissue brain transcriptomic data from ROSMAP and a similar dataset from the Mount Sinai Brain Bank showed that differentially expressed genes between the dementia and non-dementia groups were enriched for vascular-related processes (e.g., "angiogenesis") in APOE4 carriers only. Immune-related transcripts were more strongly correlated with AD pathology in APOE4 carriers with some transcripts such as TREM2 and positively correlated with pathology severity in APOE4 carriers, but negatively in non-carriers. Overall, cumulative evidence from the largest neuroimaging, pathology, and transcriptomic studies available suggests that vascular dysfunction is key to the development of AD in APOE4 carriers. However, further studies are required to tease out non-APOE4-specific mechanisms.

Identification of four serum miRNAs as potential markers to screen for thirteen cancer types

Introduction

Cancer consistently remains one of the top causes of death in the United States every year, with many cancer deaths preventable if detected early. Circulating serum miRNAs are a promising, minimally invasive supplement or even an alternative to many current screening procedures. Many studies have shown that different serum miRNAs can discriminate healthy individuals from those with certain types of cancer. Although many of those miRNAs are often reported to be significant in one cancer type, they are also altered in other cancer types. Currently, very few studies have investigated serum miRNA biomarkers for multiple cancer types for general cancer screening purposes.

Method

To identify serum miRNAs that would be useful in screening multiple types of cancers, microarray cancer datasets were curated, yielding 13 different types of cancer with a total of 3352 cancer samples and 2809 non-cancer samples. The samples were divided into training and validation sets. One hundred random forest models were built using the training set to select candidate miRNAs. The selected miRNAs were then used in the validation set to see how well they differentiate cancer from normal samples in an independent dataset. Furthermore, the interactions between these miRNAs and their target mRNAs were investigated.

Result

The random forest models achieved an average of 97% accuracy in the training set with 95% bootstrap confidence interval of 0.9544 to 0.9778. The selected miRNAs were hsa-miR-663a, hsa-miR-6802-5p, hsa-miR-6784-5p, hsa-miR-3184-5p, and hsa-miR-8073. Each miRNA exhibited high area under the curve (AUC) value using receiver operating characteristic analysis. Moreover, the combination of four out of five miRNAs achieved the highest AUC value of 0.9815 with high sensitivity of 0.9773, indicating that these miRNAs have a high potential for cancer screening. miRNA-mRNA and protein-protein interaction analysis provided insights into how these miRNAs play a role in cancer.

Overlap between Central and Peripheral Transcriptomes in Parkinson’s Disease but Not Alzheimer’s Disease

Most neurodegenerative disorders take decades to develop, and their early detection is challenged by confounding non-pathological ageing processes. Therefore, the discovery of genes and molecular pathways in both peripheral and brain tissues that are highly predictive of disease evolution is necessary. To find genes that influence Alzheimer’s disease (AD) and Parkinson’s disease (PD) pathogenesis, human RNA-Seq transcriptomic data from Brodmann Area 9 (BA9) of the dorsolateral prefrontal cortex (DLPFC), whole blood (WB), and peripheral blood mononuclear cells (PBMC) were analysed using a combination of differential gene expression and a random forest-based machine learning algorithm. The results suggest that there is little overlap between PD and AD, and the AD brain signature is unique mainly compared to blood-based samples. Moreover, the AD-BA9 was characterised by changes in ‘nervous system development’ with Myocyte-specific enhancer factor 2C (Mef2C), encoding a transcription factor that induces microglia activation, a prominent feature. The peripheral AD transcriptome was associated with alterations in ‘viral process’, and FYN, which has been previously shown to link amyloid-beta and tau, was the prominent feature. However, in the absence of any overlap with the central transcriptome, it is unclear whether peripheral FYN levels reflect AD severity or progression. In PD, central and peripheral signatures are characterised by anomalies in ‘exocytosis’ and specific genes related to the SNARE complex, including Vesicle-associated membrane protein 2 (VAMP2), Syntaxin 1A (STX1A), and p21-activated kinase 1 (PAK1). This is consistent with our current understanding of the physiological role of alpha-synuclein and how alpha-synuclein oligomers compromise vesicle docking and neurotransmission. Overall, the results describe distinct disease-specific pathomechanisms, both within the brain and peripherally, for the two most common neurodegenerative disorders.

Data-independent acquisition mass spectrometry in severe rheumatic heart disease (RHD) identifies a proteomic signature showing ongoing inflammation and effectively classifying RHD cases

BACKGROUND

Rheumatic heart disease (RHD) remains a major source of morbidity and mortality in developing countries. A deeper insight into the pathogenetic mechanisms underlying RHD could provide opportunities for drug repurposing, guide recommendations for secondary penicillin prophylaxis, and/or inform development of near-patient diagnostics.

METHODS

We performed quantitative proteomics using Sequential Windowed Acquisition of All Theoretical Fragment Ion Mass Spectrometry (SWATH-MS) to screen protein expression in 215 African patients with severe RHD, and 230 controls. We applied a machine learning (ML) approach to feature selection among the 366 proteins quantifiable in at least 40% of samples, using the Boruta wrapper algorithm. The case-control differences and contribution to Area Under the Receiver Operating Curve (AUC) for each of the 56 proteins identified by the Boruta algorithm were calculated by Logistic Regression adjusted for age, sex and BMI. Biological pathways and functions enriched for proteins were identified using ClueGo pathway analyses.

RESULTS

Adiponectin, complement component C7 and fibulin-1, a component of heart valve matrix, were significantly higher in cases when compared with controls. Ficolin-3, a protein with calcium-independent lectin activity that activates the complement pathway, was lower in cases than controls. The top six biomarkers from the Boruta analyses conferred an AUC of 0.90 indicating excellent discriminatory capacity between RHD cases and controls.

CONCLUSIONS

These results support the presence of an ongoing inflammatory response in RHD, at a time when severe valve disease has developed, and distant from previous episodes of acute rheumatic fever. This biomarker signature could have potential utility in recognizing different degrees of ongoing inflammation in RHD patients, which may, in turn, be related to prognostic severity.

Machine Learning-Based Identification of Colon Cancer Candidate Diagnostics Genes

Simple Summary

We developed a predictive approach using different machine learning methods to identify a number of genes that can potentially serve as novel diagnostic colon cancer biomarkers.

Abstract

Background: Colorectal cancer (CRC) is the third leading cause of cancer-related death and the fourth most commonly diagnosed cancer worldwide. Due to a lack of diagnostic biomarkers and understanding of the underlying molecular mechanisms, CRC’s mortality rate continues to grow. CRC occurrence and progression are dynamic processes. The expression levels of specific molecules vary at various stages of CRC, rendering its early detection and diagnosis challenging and the need for identifying accurate and meaningful CRC biomarkers more pressing. The advances in high-throughput sequencing technologies have been used to explore novel gene expression, targeted treatments, and colon cancer pathogenesis. Such approaches are routinely being applied and result in large datasets whose analysis is increasingly becoming dependent on machine learning (ML) algorithms that have been demonstrated to be computationally efficient platforms for the identification of variables across such high-dimensional datasets. Methods: We developed a novel ML-based experimental design to study CRC gene associations. Six different machine learning methods were employed as classifiers to identify genes that can be used as diagnostics for CRC using gene expression and clinical datasets. The accuracy, sensitivity, specificity, F1 score, and area under receiver operating characteristic (AUROC) curve were derived to explore the differentially expressed genes (DEGs) for CRC diagnosis. Gene ontology enrichment analyses of these DEGs were performed and predicted gene signatures were linked with miRNAs. Results: We evaluated six machine learning classification methods (Adaboost, ExtraTrees, logistic regression, naïve Bayes classifier, random forest, and XGBoost) across different combinations of training and test datasets over GEO datasets. The accuracy and the AUROC of each combination of training and test data with different algorithms were used as comparison metrics. Random forest (RF) models consistently performed better than other models. In total, 34 genes were identified and used for pathway and gene set enrichment analysis. Further mapping of the 34 genes with miRNA identified interesting miRNA hubs genes. Conclusions: We identified 34 genes with high accuracy that can be used as a diagnostics panel for CRC.

Machine Learning-Based Model for the Prognosis of Postoperative Gastric Cancer

Background

The use of machine learning (ML) in predicting disease prognosis has increased, and scientists have adopted different methods for cancer classification to optimize the early screening of cancer to determine its prognosis in advance. In this study, we aimed at improving the prediction accuracy of gastric cancer in postoperation patients by constructing a highly effective prognostic model.

Methods

The study used postoperative gastric cancer patient data from the SEER database. The LASSO regression method was used to construct a clinical prognostic model, and four machine learning methods (Boruta algorithm, neural network, support vector machine, and random forest) were used to screen and recombine the features to construct an ML prognostic model. Clinical information on 955 postoperative gastric cancer patients collected from the Affiliated Tumor Hospital of Harbin Medical University was used for external verification.

Results

Experimental results showed that the AUC values of 1, 3 and 5 years in the training set, validation set and external validation set of clinical prognosis model and ML prognosis model directly established by LASSO regression are all around 0.8.

Conclusion

Both models can accurately evaluate the prognosis of postoperative patients with gastric cancer, which may be helpful for accurate and personalized treatment of postoperative patients with gastric cancer.

Human disease biomarker panels through systems biology

As more uses for biomarkers are sought after for an increasing number of disease targets, single-target biomarkers are slowly giving way for biomarker panels. These panels incorporate various sources of biomolecular and clinical data to guarantee a higher robustness and power of separation for a clinical test. Multifactorial diseases such as psychiatric disorders show great potential for clinical use, assisting medical professionals during the analysis of risk and predisposition, disease diagnosis and prognosis, and treatment applicability and efficacy. More specific tests are also being developed to assist in ruling out, distinguishing between, and confirming suspicions of multifactorial diseases, as well as to predict which therapy option may be the best option for a given patient's biochemical profile. As more complex datasets are entering the field, involving multi-omic approaches, systems biology has stepped in to facilitate the discovery and validation steps during biomarker panel generation. Filtering biomolecules and clinical data, pre-validating and cross-validating potential biomarkers, generating final biomarker panels, and testing the robustness and applicability of those panels are all beginning to rely on machine learning and systems biology and research in this area will only benefit from advances in these approaches.

Use of Multi-Temporal LiDAR to Quantify Fertilization Effects on Stand Volume and Biomass in Late-Rotation Coastal Douglas-Fir Forests

Forest fertilization is common in coastal British Columbia as a means to increase wood production and potentially enhance carbon sequestration. Generally, the effects of fertilization are determined by measuring sample plots pre- and post-treatment, resulting in fertilization effects being determined for a limited portion of the treatment area. Applications of remote sensing-based enhanced forest inventories have allowed for estimations to expand to the wider forested area. However, these applications have not focused on monitoring the effects of silvicultural treatments. The objective of this research was to examine if a multi-temporal application of the LiDAR area-based method can be used to detect the fertilization effects on volume, biomass, and height in a second-growth Douglas-fir (Pseudotsuga menziesii) stand. The study area on Vancouver Island was fertilized in January 2007, and sample plots were established in 2011. LiDAR acquisitions were made in 2004, prior to fertilization, and in 2008, 2011, and 2016, covering both treated and untreated areas. A total of 29 paired LiDAR blocks, comprised of four 20 m resolution raster cells, were selected on either side of the fertilization boundary for analysis of the effects across several different stand types differing in the percentage of Douglas-fir, site index, and age. Random forest (RF) plot-level models were developed to estimate total stem volume and total stem biomass for each year of LiDAR acquisition using an area-based approach. Plot level results showed an increase in stem volume by 13% fertilized over control from 2005 to 2011, which was similar to a 14% increase in above-ground carbon stocks estimated using a tree-ring stand reconstruction approach. Plot-level RF models showed R2 values of 0.86 (volume) and 0.92 (biomass) with relative cross-validated root mean square errors of 12.5% (volume) and 11.9% (biomass). For both the sample plots and LiDAR blocks, statistical results indicated no significant differences in volume or biomass between treatments. However, significant differences in height increments were detected between treatments in LiDAR blocks. The results from this research highlight the promising potential for the use of enhanced forest inventory methods to rapidly expand the assessment of treatment effects beyond sample plots to the stand, block, or landscape level.