The Computational Models of Drug-target Interaction Prediction.

Drug–target interaction (DTI) prediction is a crucial step in drug discovery and repositioning as it reduces experimental validation costs if done right. Thus, developing in-silico methods to predict potential DTI has become a competitive research niche, with one of its main focuses being improving the prediction accuracy. Using machine learning (ML) models for this task, specifically network-based approaches, is effective and has shown great advantages over the other computational methods. However, ML model development involves upstream hand-crafted feature extraction and other processes that impact prediction accuracy. Thus, network-based representation learning techniques that provide automated feature extraction combined with traditional ML classifiers dealing with downstream link prediction tasks may be better-suited paradigms. Here, we present such a method, DTi2Vec, which identifies DTIs using network representation learning and ensemble learning techniques. DTi2Vec constructs the heterogeneous network, and then it automatically generates features for each drug and target using the nodes embedding technique. DTi2Vec demonstrated its ability in drug–target link prediction compared to several state-of-the-art network-based methods, using four benchmark datasets and large-scale data compiled from DrugBank. DTi2Vec showed a statistically significant increase in the prediction performances in terms of AUPR. We verified the "novel" predicted DTIs using several databases and scientific literature. DTi2Vec is a simple yet effective method that provides high DTI prediction performance while being scalable and efficient in computation, translating into a powerful drug repositioning tool.

EA-based hyperparameter optimization of hybrid deep learning models for effective drug-target interactions prediction

Central nervous system (CNS) diseases such as glioblastoma (GBM), multiple sclerosis (MS), and Alzheimer's disease (AD) remain challenging due to their complexity and limited treatments. Conventional drug repurposing strategies often rely on bulk RNA sequencing data, which can overlook cellular heterogeneity and mask rare but critical cell populations. Here, we introduce scDrugLink, a computational method that integrates single-cell transcriptomic data with drug targets and perturbation signatures to improve repurposing. For each cell type, scDrugLink constructs a Drug2Cell matrix based on drug targets to estimate promotion/inhibition scores and derives sensitivity/resistance scores by reverse matching signatures and disease-associated genes. These scores are then "linked," yielding robust therapeutic rankings. In our study, we present a systematic evaluation of single-cell drug repurposing methods for CNS diseases. Applied to atlas data for GBM, MS, and AD, scDrugLink surpassed three state-of-the-art methods (ASGARD, DrugReSC, and scDrugPrio), achieving area under the receiver operating characteristic curve (AUC) ranges of 0.6286-0.7242 and area under the precision-recall curve (AUPRC) ranges of 0.3412-0.5484. It also ranked top when comparing AUC and AUPRC at the level of individual cell types. Moreover, applying the "linking" principle to baseline methods boosted their performance, on average improving AUC and AUPRC by 0.0160 and 0.0244, respectively. Despite the advancements, the complexity and heterogeneity of CNS diseases, along with incomplete drug data, indicate that further improvement is necessary. We discuss these challenges and suggest directions for enhancing single-cell drug repurposing in the future.

Identification of interactions of drugs and proteins is an essential step in the early stages of drug discovery and in finding new drug uses. Traditional experimental identification and validation of these interactions are still time-consuming, expensive, and do not have a high success rate. To improve this identification process, development of computational methods to predict and rank likely drug-target interactions (DTI) with minimum error rate would be of great help. In this work, we propose a computational method for (Drug-Target interaction prediction using Graph Embedding and graph Mining), DTiGEM. DTiGEM models identify novel DTIs as a link prediction problem in a heterogeneous graph constructed by integrating three networks, namely: drug-drug similarity, target-target similarity, and known DTIs. DTiGEM combines different techniques, including graph embeddings (e.g., node2vec), graph mining (e.g., path scores between drugs and targets), and machine learning (e.g., different classifiers). DTiGEM achieves improvement in the prediction performance compared to other state-of-the-art methods for computational prediction of DTIs on four benchmark datasets in terms of area under precision-recall curve (AUPR). Specifically, we demonstrate that based on the average AUPR score across all benchmark datasets, DTiGEM achieves the highest average AUPR value (0.831), thus reducing the prediction error by 22.4% relative to the second-best performing method in the comparison.

BackgroundIn the pediatric intensive care unit (PICU), quantifying illness severity can be guided by risk models to enable timely identification and appropriate intervention. Logistic regression models, including the pediatric index of mortality 2 (PIM-2) and pediatric risk of mortality III (PRISM-III), produce a mortality risk score using data that are routinely available at PICU admission. Artificial neural networks (ANNs) outperform regression models in some medical fields.ObjectiveIn light of this potential, we aim to examine ANN performance, compared to that of logistic regression, for mortality risk estimation in the PICU.MethodsThe analyzed data set included patients from North American PICUs whose discharge diagnostic codes indicated evidence of infection and included the data used for the PIM-2 and PRISM-III calculations and their corresponding scores. We stratified the data set into training and test sets, with approximately equal mortality rates, in an effort to replicate real-world data. Data preprocessing included imputing missing data through simple substitution and normalizing data into binary variables using PRISM-III thresholds. A 2-layer ANN model was built to predict pediatric mortality, along with a simple logistic regression model for comparison. Both models used the same features required by PIM-2 and PRISM-III. Alternative ANN models using single-layer or unnormalized data were also evaluated. Model performance was compared using the area under the receiver operating characteristic curve (AUROC) and the area under the precision recall curve (AUPRC) and their empirical 95% CIs.ResultsData from 102,945 patients (including 4068 deaths) were included in the analysis. The highest performing ANN (AUROC 0.871, 95% CI 0.862-0.880; AUPRC 0.372, 95% CI 0.345-0.396) that used normalized data performed better than PIM-2 (AUROC 0.805, 95% CI 0.801-0.816; AUPRC 0.234, 95% CI 0.213-0.255) and PRISM-III (AUROC 0.844, 95% CI 0.841-0.855; AUPRC 0.348, 95% CI 0.322-0.367). The performance of this ANN was also significantly better than that of the logistic regression model (AUROC 0.862, 95% CI 0.852-0.872; AUPRC 0.329, 95% CI 0.304-0.351). The performance of the ANN that used unnormalized data (AUROC 0.865, 95% CI 0.856-0.874) was slightly inferior to our highest performing ANN; the single-layer ANN architecture performed poorly and was not investigated further.ConclusionsA simple ANN model performed slightly better than the benchmark PIM-2 and PRISM-III scores and a traditional logistic regression model trained on the same data set. The small performance gains achieved by this two-layer ANN model may not offer clinically significant improvement; however, further research with other or more sophisticated model designs and better imputation of missing data may be warranted.

Predicting possible drug-target interactions (DTIs) involves estimating the likelihood that drug molecules would bind to biological targets. By looking into the high cost and risk of in vivo and in vitro experiments, computational methods offer a cost- and time-efficient alternative for predicting DTIs in drug discovery. Many network-based methods have made significant progress in DTI prediction, but they often lack the ability to capture the topological and biological closeness between nodes, which is a crucial paradigm to identify hidden patterns in the DTI network. In this paper, we propose a novel computational-based framework─CAHNetF-DTP (Community Aware Heterogeneous Network-based Framework for Drug-Target interaction Prediction) that leverages a heterogeneous biomedical network comprising 15 distinct types of similarity-based subnetworks derived from drugs, targets, diseases, and side effects. To effectively capture the complex topological and semantic relationships among entities, we employ a community-aware framework that helps to capture local interaction patterns first before moving to another community. The resulting sequences are treated like sentences, enabling Word2Vec to generate embeddings that capture the contextual information on biomedical entities across multiple biological dimensions. To address the class imbalance issue, we incorporate a controlled negative sampling strategy and evaluate the proposed model under five experimental settings. The proposed method is further validated on benchmark data sets (KIBA and DAVIS), in real-world scenarios, and against SAR-based target prediction tools, where CAHNetF-DTP demonstrates robust and competitive performance, outperforming most of the methods. We carried out various case studies to check the model's capability to identify novel drug-target pairs. This work highlights the value of integrating heterogeneous biological data for improved DTI prediction.

IntroductionThe COVID-19 pandemic has exacerbated mental health challenges, particularly depression among college students. Detecting at-risk students early is crucial but remains challenging, particularly in developing countries. Utilizing data-driven predictive models presents a viable solution to address this pressing need.Aims1) To develop and compare machine learning (ML) models for predicting depression in Argentinean students during the pandemic. 2) To assess the performance of classification and regression models using appropriate metrics. 3) To identify key features driving depression prediction.MethodsA longitudinal dataset (N = 1492 college students) captured T1 and T2 measurements during the Argentinean COVID-19 quarantine. ML models, including linear logistic regression classifiers/ridge regression (LogReg/RR), random forest classifiers/regressors, and support vector machines/regressors (SVM/SVR), are employed. Assessed features encompass depression and anxiety scores (at T1), mental disorder/suicidal behavior history, quarantine sub-period information, sex, and age. For classification, models’ performance on test data is evaluated using Area Under the Precision-Recall Curve (AUPRC), Area Under the Receiver Operating Characteristic curve, Balanced Accuracy, F1 score, and Brier loss. For regression, R-squared (R2), Mean Absolute Error, and Mean Squared Error are assessed. Univariate analyses are conducted to assess the predictive strength of each individual feature with respect to the target variable. The performance of multi- vs univariate models is compared using the mean AUPRC score for classifiers and the R2 score for regressors.ResultsThe highest performance is achieved by SVM and LogReg (e.g., AUPRC: 0.76, 95% CI: 0.69, 0.81) and SVR and RR models (e.g., R2 for SVR and RR: 0.56, 95% CI: 0.45, 0.64 and 0.45, 0.63, respectively). Univariate models, particularly LogReg and SVM using depression (AUPRC: 0.72, 95% CI: 0.64, 0.79) or anxiety scores (AUPRC: 0.71, 95% CI: 0.64, 0.78) and RR using depression scores (R2: 0.48, 95% CI: 0.39, 0.57) exhibit performance levels close to those of the multivariate models, which include all features.DiscussionThese findings highlight the relevance of pre-existing depression and anxiety conditions in predicting depression during quarantine, underscoring their comorbidity. ML models, particularly SVM/SVR and LogReg/RR, demonstrate potential in the timely detection of at-risk students. However, further studies are needed before clinical implementation.

Random Forest Modeling of Acute Toxicity in Anal Cancer: Effects of Peritoneal Cavity Contouring Approaches on Model Performance

To evaluate the performance of different probabilistic classifiers to predict posterior capsule rupture (PCR) prior to cataract surgery. Three probabilistic classifiers were constructed to estimate the probability of PCR: a Bayesian network (BN), logistic regression (LR) model, and multi-layer perceptron (MLP) network. The classifiers were trained on a sample of 2853 376 surgeries reported to the European Registry of Quality Outcomes for Cataract and Refractive Surgery (EUREQUO) between 2008 and 2018. The performance of the classifiers was evaluated based on the area under the precision-recall curve (AUPRC) and compared to existing scoring models in the literature. Furthermore, direct risk factors for PCR were identified by analysing the independence structure of the BN. The MLP network predicted PCR overall the best (AUPRC 13.1 ± 0.41%), followed by the BN (AUPRC 8.05 ± 0.39%) and the LR model (AUPRC 7.31 ± 0.15%). Direct risk factors for PCR include preoperative best-corrected visual acuity (BCVA), year of surgery, operation type, anaesthesia, target refraction, other ocular comorbidities, white cataract, and corneal opacities. Our results suggest that the MLP network performs better than existing scoring models in the literature, despite a relatively low precision at high recall. Consequently, implementing the MLP network in clinical practice can potentially decrease the PCR rate.

Electronic health records (EHRs) with clinical decision support tools are now ubiquitous in healthcare organizations. Clinical foundation models (CFMs) pretrained on large-scale, heterogeneous structured EHR data have emerged as a powerful approach to improve predictive performance and generalizability. Meanwhile, large language models (LLMs) pretrained on broad data sources are being applied to an expanding range of healthcare tasks. However, it remains unclear whether generalist LLMs can match specialized CFMs for disease risk prediction using structured clinical data. We compared CFMs (Med-BERT, CLMBR) against fine-tuned generalist LLMs (Mistral, LLaMA-2/3/3.1), a clinical LLM (Me-LLaMA), and LLM-generated embeddings paired with simple classifiers (using DeepSeek, Qwen3, and GPT-OSS) on two disease risk prediction tasks: heart failure risk among diabetic patients (DHF) and pancreatic cancer diagnosis (PaCa). Evaluations spanned multi-site EHR data, claims data, and an open-source single-institution benchmark (EHRSHOT). Performance was assessed using the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC). On larger EHR and claims cohorts (>30,000 patients), fine-tuned CFMs outperformed fine-tuned LLMs by a small but statistically significant margin (<1% AUROC). The clinical LLM performed comparably to generalist LLMs despite being smaller. On the open-source PaCa cohort (3,810 patients, 199 cases), LLMs achieved slightly higher AUROCs that were not statistically significant (LLaMA-3.1-70B 86.1% vs. Med-BERT 85.3%, p=0.27), but CFMs achieved significantly higher AUPRC (Med-BERT 55.9% vs. LLaMA-3.1-70B 41.1%, p=0.001). Notably, LLM-generated trajectory embeddings paired with logistic regression or a simple MLP, without any LLM fine-tuning, achieved the best overall performance, with AUROC exceeding 90% (Qwen3) and AUPRC reaching 66% (GPT-OSS 20B). LLM-generated embeddings with lightweight classifiers outperformed both fine-tuned CFMs and fine-tuned LLMs on AUROC and AUPRC. While these results demonstrate the potential of generalist models to match or surpass specialized CFMs, their substantially greater computational cost and variable AUPRC performance in the fine-tuning setting warrant caution. We provide a reproducible evaluation framework and codebase to support continued benchmarking.

Coronary artery disease (CAD) results in substantial morbidity and mortality. The purpose of this study was to develop a deep learning model to detect CAD defined using diagnostic codes ("ECG2CAD") and identify people at risk for adverse events using electrocardiograms (ECGs) in a primary care setting. ECG2CAD was trained on 764,670 ECGs representing 137,199 individuals at Massachusetts General Hospital (MGH). Model performance for discrimination of prevalent CAD was measured using area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC), and compared against model of age and sex, and Pooled Cohort Equations, in 3 test sets: MGH, Brigham and Women's Hospital (BWH), and UK Biobank. Subgroups were assessed for incident CAD-related events in a BWH primary care cohort. ECG2CAD was evaluated in MGH (N = 18,706 [6,051 cases], age 57 ± 16 years), BWH (N = 88,270 [27,898 cases], age 57 ± 16 years), and UK Biobank (N = 42,147 [1,509 cases], age 65 ± 8 years). ECG2CAD consistently discriminated prevalent CAD (MGH AUROC: 0.782; AUPRC: 0.639; BWH: AUROC: 0.747; AUPRC: 0.588; UK Biobank AUROC: 0.760; AUPRC: 0.155) and incrementally vs models based on age and sex or Pooled Cohort Equations (P < 0.01) in MGH and BWH. In the BWH primary care subset, model performance was consistent across subgroups. Being in the highest quintile of ECG2CAD risk was associated with higher risk for adverse events compared with low-risk group (myocardial infarction HR: 5.59; 95% CI: 4.76-6.56, heart failure 10.49; 95% CI: 7.96-13.84, all-cause mortality 2.68; 95% CI: 2.32-3.10). Artificial intelligence-enabled analysis of the ECG may facilitate identification of individuals with possible undiagnosed CAD and inform downstream testing and preventive measures.

Drug-target interaction identification is an essential step of drug discovering and adverse effect prediction. In real bio-environment, the connections among drugs and targets, as well as themselves construct a complex heterogeneous network, and profoundly affect the predictive performance of drug-target interactions. However, the current methods usually focus on the drug-target interactions alone, which may be very sparse and with numerous noises, may not produce satisfactory prediction results. In this paper, we propose a novel approach, dual-graph regularized robust PCA in heterogeneous network, for drug-target interaction prediction task. In particular, we aim at decompose the bipartite graph of drug-target interactions into two low-rank matrices, which represent the latent representations of drugs and targets respectively, and smooth the drug-drug and target-target graphs simultaneously. Moreover, an improved robust PCA model is used to suppress the widespread noisy connections in the decomposition stage. For the optimization, we design an efficient algorithm to solve few subproblems with close-form solution. Finally the extensive experiments on real world drug-target heterogeneous networks are presented to show the effectiveness of the proposed methods.

Simple SummaryThe identification of cancer driver genes is, for statistical reasons, often biased toward genes that are altered frequently in a cohort. However, genes that are less frequently mutated can also alter cancer hallmarks. To detect such rarely mutated genes involved in driving metastatic prostate cancer, we analyzed the Hartwig Medical Foundation metastatic prostate cancer cohort. Hereto, we developed GoNetic, a novel network-based method that can detect genes with a lower mutational rate as members of recurrently mutated sets of genes connected on a prior interaction network. In contrast to state-of-the-art network-based driver identification methods, GoNetic retains information on sample-specific mutations and uses more properties of the prior interaction network. When applied to the Hartwig Medical Foundation cohort, GoNetic successfully prioritized both known drivers and rarely mutated driver candidates of metastatic prostate cancer. Comprehensive validation with other public data sets further supported the driver potential of these novel candidates.Most known driver genes of metastatic prostate cancer are frequently mutated. To dig into the long tail of rarely mutated drivers, we performed network-based driver identification on the Hartwig Medical Foundation metastatic prostate cancer data set (HMF cohort). Hereto, we developed GoNetic, a method based on probabilistic pathfinding, to identify recurrently mutated subnetworks. In contrast to most state-of-the-art network-based methods, GoNetic can leverage sample-specific mutational information and the weights of the underlying prior network. When applied to the HMF cohort, GoNetic successfully recovered known primary and metastatic drivers of prostate cancer that are frequently mutated in the HMF cohort (TP53, RB1, and CTNNB1). In addition, the identified subnetworks contain frequently mutated genes, reflect processes related to metastatic prostate cancer, and contain rarely mutated driver candidates. To further validate these rarely mutated genes, we assessed whether the identified genes were more mutated in metastatic than in primary samples using an independent cohort. Then we evaluated their association with tumor evolution and with the lymph node status of the patients. This resulted in forwarding several novel putative driver genes for metastatic prostate cancer, some of which might be prognostic for disease evolution.

Hospital-onset bacteremia and fungemia (HOB) are common and potentially preventable complications of hospital care. To assess whether nonindividual patient features, which summarize interactions with other patients and health care workers (HCWs), can contribute to predictive and causal machine learning models for HOB. This prognostic study included adult patients admitted to Barnes-Jewish Hospital, an academic hospital in St Louis, Missouri, in 2021. Analyses were developed between October 2023 and August 2024 and in April 2025. Individual patient features were extracted from electronic health records and used to engineer nonpatient features, including interactions with HCWs and direct or indirect (consecutive room occupancy) patient contact. HOB was defined as a positive blood culture after the third day of hospitalization. Patients who were hospitalized for more than 3 days were considered at risk for the outcome. We developed 3 gradient boosting models: 2 predictive (with patient features only and with both patient and nonpatient features to predict the occurrence of HOB) and 1 causal to test the association of nonpatient features and HOB. Predictive performance is reported using area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC), and the results of the causal model are reported as difference in average effects. Sensitivity analyses separated intensive care unit-onset and ward-onset HOB and included a methicillin-resistant Staphylococcus aureus-specific model to adjust for colonization pressure. Among the 52 442 patients, 34 855 (66.5%) had admissions longer than 72 hours and were included for analysis; of these, 556 (1.6%) developed HOB. The median age for the included patients was 60 (IQR, 44-70) years, 50.5% were female, and obesity was the most frequent comorbidity (25.0%). Nonpatient features, such as a prior occupant of the same room receiving antipseudomonal beta-lactams and the mean number of HCWs per day for the 7 days preceding HOB, improved the model's performance (AUROC, 0.88 [95% CI, 0.88-0.89]; AUPRC, 0.20 [95% CI, 0.20-0.22]) compared with the patient-only model (AUROC, 0.85 [95% CI, 0.85-0.86]; AUPRC, 0.13 [95% CI, 0.12-0.14]) (P < .001). These 2 features were also associated with a higher likelihood of HOB in the causal gradient boosting model. These findings suggest that nonindividual patient features may contribute to a comprehensive analysis of HOB when integrated with individual patient features in a machine learning model.

The World Health Organization glioma classification highlights genetic profiles, such as isocitrate dehydrogenase (IDH) mutation. This study developed an uncertainty-aware deep learning model to predict IDH mutations in glioma patients, employing conformal prediction (CP) for uncertainty quantification (UQ). The UCSF dataset split into training (70%) and validation (30%) sets, and the UPENN dataset divided into calibration (30%) and external test (70%) sets. We developed various 3D convolutional neural network models and selected the best based on the validation set area under the precision-recall curve (AUPRC). Also, we trained a logistic regression (LR) ensemble classifier on the training set and selected the best ensemble model based on the validation set AUPRC. Finally, we employed CP, with nonconformity thresholds set at the error rate of 0.01. The conformal model was calibrated on the calibration set and was evaluated on the external test set using the AUPRC and the area under the receiver operating characteristic (AUROC). 3D convolutional models achieved a mean AUROC and AUPRC of 0.9004 and 0.8102 on the validation set, and a mean AUROC and AUPRC of 0.7340 and 0.2139 on the external test set, respectively. Using the LR ensemble model, the model achieved an AUROC and AUPRC of 0.79820 and 0.2583 on the external test set, respectively. CP with a 0.01 nonconformity threshold with 0.9917 coverage reached an AUROC and AUPRC of 0.8592 and 0.5217, respectively. Integrating CP for UQ improves the performance of deep learning models for predicting IDH mutation in glioma patients.