On the road to explainable AI in drug-drug interactions prediction: a systematic review

BiRNN-DDI: A Drug-Drug Interaction Event Type Prediction Model Based on Bidirectional Recurrent Neural Network and Graph2Seq Representation

Research on drug-drug interaction (DDI) prediction, particularly in identifying DDI event types, is crucial for understanding adverse drug reactions and drug combinations. This work introduces a Bidirectional Recurrent Neural Network model for DDI event type prediction (BiRNN-DDI), which simultaneously considers structural relationships and contextual information. Our BiRNN-DDI model constructs drug feature graphs to mine structural relationships. For contextual information, it transforms drug graphs into sequences and employs a two-channel structure, integrating BiRNN, to obtain contextual representations of drug-drug pairs. The model's effectiveness is demonstrated through comparisons with state-of-the-art models on two DDI event-type benchmarks. Extensive experimental results reveal that BiRNN-DDI surpasses other models in accuracy, AUPR, AUC, F1 score, Precision, and Recall metrics on both small and large datasets. Additionally, our model exhibits a lower parameter space, indicating more efficient learning of drug feature representations and prediction of potential DDI event types.

Recent progress in machine learning approaches for predicting carcinogenicity in drug development

INTRODUCTION

This review explores the transformative impact of machine learning (ML) on carcinogenicity prediction within drug development. It discusses the historical context and recent advancements, emphasizing the significance of ML methodologies in overcoming challenges related to data interpretation, ethical considerations, and regulatory acceptance.

AREAS COVERED

The review comprehensively examines the integration of ML, deep learning, and diverse artificial intelligence (AI) approaches in various aspects of drug development safety assessments. It explores applications ranging from early-phase compound screening to clinical trial optimization, highlighting the versatility of ML in enhancing predictive accuracy and efficiency.

EXPERT OPINION

Through the analysis of traditional approaches such as in vivo rodent bioassays and in vitro assays, the review underscores the limitations and resource intensity associated with these methods. It provides expert insights into how ML offers innovative solutions to address these challenges, revolutionizing safety assessments in drug development.

New methods for drug synergy prediction: A mini-review

In this mini-review, we explore the new prediction methods for drug combination synergy relying on high-throughput combinatorial screens. The fast progress of the field is witnessed in the more than thirty original machine learning methods published since 2021, a clear majority of them based on deep learning techniques. We aim to put these articles under a unifying lens by highlighting the core technologies, the data sources, the input data types and synergy scores used in the methods, as well as the prediction scenarios and evaluation protocols that the articles deal with. Our finding is that the best methods accurately solve the synergy prediction scenarios involving known drugs or cell lines while the scenarios involving new drugs or cell lines still fall short of an accurate prediction level.

Application of Artificial Intelligence in Drug-Drug Interactions Prediction: A Review

Drug-drug interactions (DDI) are a critical aspect of drug research that can have adverse effects on patients and can lead to serious consequences. Predicting these events accurately can significantly improve clinicians' ability to make better decisions and establish optimal treatment regimens. However, manually detecting these interactions is time-consuming and labor-intensive. Utilizing the advancements in Artificial Intelligence (AI) is essential for achieving accurate forecasts of DDIs. In this review, DDI prediction tasks are classified into three types according to the type of DDI prediction: undirected DDI prediction, DDI events prediction, and Asymmetric DDI prediction. The paper then reviews the progress of AI for each of these three prediction tasks in DDI and provides a summary of the data sets used as well as the representative methods used in these three prediction directions. In this review, we aim to provide a comprehensive overview of drug interaction prediction. The first section introduces commonly used databases and presents an overview of current research advancements and techniques across three domains of DDI. Additionally, we introduce classical machine learning techniques for predicting undirected drug interactions and provide a timeline for the progression of the predicted drug interaction events. At last, we debate the difficulties and prospects of AI approaches at predicting DDI, emphasizing their potential for improving clinical decision-making and patient outcomes.

Drug-drug interactions prediction based on deep learning and knowledge graph: A review

Drug-drug interactions (DDIs) can produce unpredictable pharmacological effects and lead to adverse events that have the potential to cause irreversible damage to the organism. Traditional methods to detect DDIs through biological or pharmacological analysis are time-consuming and expensive, therefore, there is an urgent need to develop computational methods to effectively predict drug-drug interactions. Currently, deep learning and knowledge graph techniques which can effectively extract features of entities have been widely utilized to develop DDI prediction methods. In this research, we aim to systematically review DDI prediction researches applying deep learning and graph knowledge. The available biomedical data and public databases related to drugs are firstly summarized in this review. Then, we discuss the existing drug-drug interactions prediction methods which have utilized deep learning and knowledge graph techniques and group them into three main classes: deep learning-based methods, knowledge graph-based methods, and methods that combine deep learning with knowledge graph. We comprehensively analyze the commonly used drug related data and various DDI prediction methods, and compare these prediction methods on benchmark datasets. Finally, we briefly discuss the challenges related to drug-drug interactions prediction, including asymmetric DDIs prediction and high-order DDI prediction.

Integrating Artificial Intelligence for Drug Discovery in the Context of Revolutionizing Drug Delivery

Drug development is expensive, time-consuming, and has a high failure rate. In recent years, artificial intelligence (AI) has emerged as a transformative tool in drug discovery, offering innovative solutions to complex challenges in the pharmaceutical industry. This manuscript covers the multifaceted role of AI in drug discovery, encompassing AI-assisted drug delivery design, the discovery of new drugs, and the development of novel AI techniques. We explore various AI methodologies, including machine learning and deep learning, and their applications in target identification, virtual screening, and drug design. This paper also discusses the historical development of AI in medicine, emphasizing its profound impact on healthcare. Furthermore, it addresses AI's role in the repositioning of existing drugs and the identification of drug combinations, underscoring its potential in revolutionizing drug delivery systems. The manuscript provides a comprehensive overview of the AI programs and platforms currently used in drug discovery, illustrating the technological advancements and future directions of this field. This study not only presents the current state of AI in drug discovery but also anticipates its future trajectory, highlighting the challenges and opportunities that lie ahead.

Learning self-supervised molecular representations for drug-drug interaction prediction

Drug-drug interactions (DDI) are a critical concern in healthcare due to their potential to cause adverse effects and compromise patient safety. Supervised machine learning models for DDI prediction need to be optimized to learn abstract, transferable features, and generalize to larger chemical spaces, primarily due to the scarcity of high-quality labeled DDI data. Inspired by recent advances in computer vision, we present SMR-DDI, a self-supervised framework that leverages contrastive learning to embed drugs into a scaffold-based feature space. Molecular scaffolds represent the core structural motifs that drive pharmacological activities, making them valuable for learning informative representations. Specifically, we pre-trained SMR-DDI on a large-scale unlabeled molecular dataset. We generated augmented views for each molecule via SMILES enumeration and optimized the embedding process through contrastive loss minimization between views. This enables the model to capture relevant and robust molecular features while reducing noise. We then transfer the learned representations for the downstream prediction of DDI. Experiments show that the new feature space has comparable expressivity to state-of-the-art molecular representations and achieved competitive DDI prediction results while training on less data. Additional investigations also revealed that pre-training on more extensive and diverse unlabeled molecular datasets improved the model's capability to embed molecules more effectively. Our results highlight contrastive learning as a promising approach for DDI prediction that can identify potentially hazardous drug combinations using only structural information.

Comprehensive Review of Drug-Drug Interaction Prediction Based on Machine Learning: Current Status, Challenges, and Opportunities

Detecting drug-drug interactions (DDIs) is an essential step in drug development and drug administration. Given the shortcomings of current experimental methods, the machine learning (ML) approach has become a reliable alternative, attracting extensive attention from the academic and industrial fields. With the rapid development of computational science and the growing popularity of cross-disciplinary research, a large number of DDI prediction studies based on ML methods have been published in recent years. To give an insight into the current situation and future direction of DDI prediction research, we systemically review these studies from three aspects: (1) the classic DDI databases, mainly including databases of drugs, side effects, and DDI information; (2) commonly used drug attributes, which focus on chemical, biological, and phenotypic attributes for representing drugs; (3) popular ML approaches, such as shallow learning-based, deep learning-based, recommender system-based, and knowledge graph-based methods for DDI detection. For each section, related studies are described, summarized, and compared, respectively. In the end, we conclude the research status of DDI prediction based on ML methods and point out the existing issues, future challenges, potential opportunities, and subsequent research direction.

Leveraging transformers-based language models in proteome bioinformatics

In recent years, the rapid growth of biological data has increased interest in using bioinformatics to analyze and interpret this data. Proteomics, which studies the structure, function, and interactions of proteins, is a crucial area of bioinformatics. Using natural language processing (NLP) techniques in proteomics is an emerging field that combines machine learning and text mining to analyze biological data. Recently, transformer-based NLP models have gained significant attention for their ability to process variable-length input sequences in parallel, using self-attention mechanisms to capture long-range dependencies. In this review paper, we discuss the recent advancements in transformer-based NLP models in proteome bioinformatics and examine their advantages, limitations, and potential applications to improve the accuracy and efficiency of various tasks. Additionally, we highlight the challenges and future directions of using these models in proteome bioinformatics research. Overall, this review provides valuable insights into the potential of transformer-based NLP models to revolutionize proteome bioinformatics.

Artificial intelligence for neurodegenerative experimental models

INTRODUCTION

Experimental models are essential tools in neurodegenerative disease research. However, the translation of insights and drugs discovered in model systems has proven immensely challenging, marred by high failure rates in human clinical trials.

METHODS

Here we review the application of artificial intelligence (AI) and machine learning (ML) in experimental medicine for dementia research.

RESULTS

Considering the specific challenges of reproducibility and translation between other species or model systems and human biology in preclinical dementia research, we highlight best practices and resources that can be leveraged to quantify and evaluate translatability. We then evaluate how AI and ML approaches could be applied to enhance both cross-model reproducibility and translation to human biology, while sustaining biological interpretability.

DISCUSSION

AI and ML approaches in experimental medicine remain in their infancy. However, they have great potential to strengthen preclinical research and translation if based upon adequate, robust, and reproducible experimental data.

HIGHLIGHTS

There are increasing applications of AI in experimental medicine. We identified issues in reproducibility, cross-species translation, and data curation in the field. Our review highlights data resources and AI approaches as solutions. Multi-omics analysis with AI offers exciting future possibilities in drug discovery.

Drug-drug interactions between COVID-19 therapeutics and psychotropic medications

INTRODUCTION

The coronavirus (COVID-19) pandemic has led to as well as exacerbated mental health disorders, leading to increased use of psychotropic medications. Co-administration of COVID-19 and psychotropic medications may result in drug-drug interactions (DDIs), that may compromise both the safety and efficacy of both medications.

AREAS COVERED

This review provides an update of the current evidence on DDIs between COVID-19 and psychotropic medications. The interactions are categorized into pharmacokinetic, pharmacodynamic, and other relevant types. A thorough literature search was conducted using electronic databases to identify relevant studies, and extract data to highlight potential DDIs, clinical implications, and management strategies.

EXPERT OPINION

Understanding and managing potential DDIs between COVID-19 and psychotropic medications is paramount to ensuring safe and effective treatment of patients with COVID-19 and mental illness. Awareness of the diverse spectrum of DDIs, vigilant monitoring, and judicious dose modifications, while choosing pharmacotherapeutic options with low risk of interaction whenever possible, are necessary. Ongoing and future investigations should continue to review the dynamic landscape of COVID-19 therapeutic modalities and their interactions with psychotropic medications.

MATT-DDI: Predicting multi-type drug-drug interactions via heterogeneous attention mechanisms

The joint use of multiple drugs can result in adverse drug-drug interactions (DDIs) and side effects that harm the body. Accurate identification of DDIs is crucial for avoiding accidental drug side effects and understanding potential mechanisms underlying DDIs. Several computational methods have been proposed for multi-type DDI prediction, but most rely on the similarity profiles of drugs as the drug feature vectors, which may result in information leakage and overoptimistic performance when predicting interactions between new drugs. To address this issue, we propose a novel method, MATT-DDI, for predicting multi-type DDIs based on the original feature vectors of drugs and multiple attention mechanisms. MATT-DDI consists of three main modules: the top k most similar drug pair selection module, heterogeneous attention mechanism module and multi‑type DDI prediction module. Firstly, based on the feature vector of the input drug pair (IDP), k drug pairs that are most similar to the input drug pair from the training dataset are selected according to cosine similarity between drug pairs. Then, the vectors of k selected drug pairs are averaged to obtain a new drug pair (NDP). Next, IDP and NDP are fed into heterogeneous attention modules, including scaled dot product attention and bilinear attention, to extract latent feature vectors. Finally, these latent feature vectors are taken as input of the classification module to predict DDI types. We evaluated MATT-DDI on three different tasks. The experimental results show that MATT-DDI provides better or comparable performance compared to several state-of-the-art methods, and its feasibility is supported by case studies. MATT-DDI is a robust model for predicting multi-type DDIs with excellent performance and no information leakage.

Carbon price prediction based on decomposition technique and extreme gradient boosting optimized by the grey wolf optimizer algorithm

It is essential to predict carbon prices precisely in order to reduce CO2 emissions and mitigate global warming. As a solution to the limitations of a single machine learning model that has insufficient forecasting capability in the carbon price prediction problem, a carbon price prediction model (GWO-XGBOOST-CEEMDAN) based on the combination of grey wolf optimizer (GWO), extreme gradient boosting (XGBOOST), and complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is put forward in this paper. First, a random forest (RF) method is employed to screen the primary carbon price indicators and determine the main influencing factors. Second, the GWO-XGBOOST model is established, and the GWO algorithm is utilized to optimize the XGBOOST model parameters. Finally, the residual series of the GWO-XGBOOST model are decomposed and corrected using the CEEMDAN method to produce the GWO-XGBOOST-CEEMDAN model. Three carbon emission trading markets, Guangdong, Hubei, and Fujian, were experimentally predicted to verify the model's validity. Based on the experimental results, it has been demonstrated that the proposed hybrid model has enhanced prediction precision compared to the comparison model, providing an effective experimental method for the prediction of future carbon prices.

Identification of S1PR4 as an immune modulator for favorable prognosis in HNSCC through machine learning

G protein-coupled receptors (GPCRs) are the largest family of membrane proteins and play a critical role as pharmacological targets. An improved understanding of GPCRs' involvement in tumor microenvironment may provide new perspectives for cancer therapy. This study used machine learning to classify head and neck squamous cell carcinoma (HNSCC) patients into two GPCR-based subtypes. Notably, these subtypes showed significant differences in prognosis, gene expression, and immune microenvironment, particularly CD8+ T cell infiltration. S1PR4 emerged as a key regulator distinguishing the subtypes, positively correlated with CD8+ T cell proportion and cytotoxicity in HNSCC. It was predominantly expressed in CX3CR1+CD8+ T cells among T cells. Upregulation of S1PR4 enhanced T cell function during CAR-T cell therapy, suggesting its potential in cancer immunotherapy. These findings highlight S1PR4 as an immune modulator for favorable prognosis in HNSCC, and offer a potential GPCR-targeted therapeutic option for HNSCC treatment.

Identification of microbial markers associated with lung cancer based on multi-cohort 16 s rRNA analyses: A systematic review and meta-analysis

BACKGROUND

The relationship between commensal microbiota and lung cancer (LC) has been studied extensively. However, developing replicable microbiological markers for early LC diagnosis across multiple populations has remained challenging. Current studies are limited to a single region, single LC subtype, and small sample size. Therefore, we aimed to perform the first large-scale meta-analysis for identifying micro biomarkers for LC screening by integrating gut and respiratory samples from multiple studies and building a machine-learning classifier.

METHODS

In total, 712 gut and 393 respiratory samples were assessed via 16 s rRNA amplicon sequencing. After identifying the taxa of differential biomarkers, we established random forest models to distinguish between LC populations and normal controls. We validated the robustness and specificity of the model using external cohorts. Moreover, we also used the KEGG database for the predictive analysis of colony-related functions.

RESULTS

The α and β diversity indices indicated that LC patients' gut microbiota (GM) and lung microbiota (LM) differed significantly from those of the healthy population. Linear discriminant analysis (LDA) of effect size (LEfSe) helped us identify the top-ranked biomarkers, Enterococcus, Lactobacillus, and Escherichia, in two microbial niches. The area under the curve values of the diagnostic model for the two sites were 0.81 and 0.90, respectively. KEGG enrichment analysis also revealed significant differences in microbiota-associated functions between cancer-affected and healthy individuals that were primarily associated with metabolic disturbances.

CONCLUSIONS

GM and LM profiles were significantly altered in LC patients, compared to healthy individuals. We identified the taxa of biomarkers at the two loci and constructed accurate diagnostic models. This study demonstrates the effectiveness of LC-specific microbiological markers in multiple populations and contributes to the early diagnosis and screening of LC.

Explanations as a New Metric for Feature Selection: A Systematic Approach

With the extensive use of Machine Learning (ML) in the biomedical field, there was an increasing need for Explainable Artificial Intelligence (XAI) to improve transparency and reveal complex hidden relationships between variables for medical practitioners, while meeting regulatory requirements. Feature Selection (FS) is widely used as a part of a biomedical ML pipeline to significantly reduce the number of variables while preserving as much information as possible. However, the choice of FS methods affects the entire pipeline including the final prediction explanations, whereas very few works investigate the relationship between FS and model explanations. Through a systematic workflow performed on 145 datasets and an illustration on medical data, the present work demonstrated the promising complementarity of two metrics based on explanations (using ranking and influence changes) in addition to accuracy and retention rate to select the most appropriate FS/ML models. Measuring how much explanations differ with/without FS are particularly promising for FS methods recommendation. While reliefF generally performs the best on average, the optimal choice may vary for each dataset. Positioning FS methods in a tridimensional space, integrating explanations-based metrics, accuracy and retention rate, would allow the user to choose the priorities to be given on each of the dimensions. In biomedical applications, where each medical condition may have its own preferences, this framework will make it possible to offer the healthcare professional the appropriate FS technique, to select the variables that have an important explainable impact, even if this comes at the expense of a limited drop of accuracy.

Structural interaction fingerprints and machine learning for predicting and explaining binding of small molecule ligands to RNA

Ribonucleic acids (RNAs) play crucial roles in living organisms and some of them, such as bacterial ribosomes and precursor messenger RNA, are targets of small molecule drugs, whereas others, e.g. bacterial riboswitches or viral RNA motifs are considered as potential therapeutic targets. Thus, the continuous discovery of new functional RNA increases the demand for developing compounds targeting them and for methods for analyzing RNA-small molecule interactions. We recently developed fingeRNAt-a software for detecting non-covalent bonds formed within complexes of nucleic acids with different types of ligands. The program detects several non-covalent interactions and encodes them as structural interaction fingerprint (SIFt). Here, we present the application of SIFts accompanied by machine learning methods for binding prediction of small molecules to RNA. We show that SIFt-based models outperform the classic, general-purpose scoring functions in virtual screening. We also employed Explainable Artificial Intelligence (XAI)-the SHapley Additive exPlanations, Local Interpretable Model-agnostic Explanations and other methods to help understand the decision-making process behind the predictive models. We conducted a case study in which we applied XAI on a predictive model of ligand binding to human immunodeficiency virus type 1 trans-activation response element RNA to distinguish between residues and interaction types important for binding. We also used XAI to indicate whether an interaction has a positive or negative effect on binding prediction and to quantify its impact. Our results obtained using all XAI methods were consistent with the literature data, demonstrating the utility and importance of XAI in medicinal chemistry and bioinformatics.

Explainability and white box in drug discovery

Recently, artificial intelligence (AI) techniques have been increasingly used to overcome the challenges in drug discovery. Although traditional AI techniques generally have high accuracy rates, there may be difficulties in explaining the decision process and patterns. This can create difficulties in understanding and making sense of the outputs of algorithms used in drug discovery. Therefore, using explainable AI (XAI) techniques, the causes and consequences of the decision process are better understood. This can help further improve the drug discovery process and make the right decisions. To address this issue, Explainable Artificial Intelligence (XAI) emerged as a process and method that securely captures the results and outputs of machine learning (ML) and deep learning (DL) algorithms. Using techniques such as SHAP (SHApley Additive ExPlanations) and LIME (Locally Interpretable Model-Independent Explanations) has made the drug targeting phase clearer and more understandable. XAI methods are expected to reduce time and cost in future computational drug discovery studies. This review provides a comprehensive overview of XAI-based drug discovery and development prediction. XAI mechanisms to increase confidence in AI and modeling methods. The limitations and future directions of XAI in drug discovery are also discussed.

Evaluation of nutritional status and clinical depression classification using an explainable machine learning method

Introduction

Depression is a prevalent disorder worldwide, with potentially severe implications. It contributes significantly to an increased risk of diseases associated with multiple risk factors. Early accurate diagnosis of depressive symptoms is a critical first step toward management, intervention, and prevention. Various nutritional and dietary compounds have been suggested to be involved in the onset, maintenance, and severity of depressive disorders. Despite the challenges to better understanding the association between nutritional risk factors and the occurrence of depression, assessing the interplay of these markers through supervised machine learning remains to be fully explored.

Methods

This study aimed to determine the ability of machine learning-based decision support methods to identify the presence of depression using publicly available health data from the Korean National Health and Nutrition Examination Survey. Two exploration techniques, namely, uniform manifold approximation and projection and Pearson correlation, were performed for explanatory analysis among datasets. A grid search optimization with cross-validation was performed to fine-tune the models for classifying depression with the highest accuracy. Several performance measures, including accuracy, precision, recall, F1 score, confusion matrix, areas under the precision-recall and receiver operating characteristic curves, and calibration plot, were used to compare classifier performances. We further investigated the importance of the features provided: visualized interpretation using ELI5, partial dependence plots, and local interpretable using model-agnostic explanations and Shapley additive explanation for the prediction at both the population and individual levels.

Results

The best model achieved an accuracy of 86.18% for XGBoost and an area under the curve of 84.96% for the random forest model in original dataset and the XGBoost algorithm with an accuracy of 86.02% and an area under the curve of 85.34% in the quantile-based dataset. The explainable results revealed a complementary observation of the relative changes in feature values, and, thus, the importance of emergent depression risks could be identified.

Discussion

The strength of our approach is the large sample size used for training with a fine-tuned model. The machine learning-based analysis showed that the hyper-tuned model has empirically higher accuracy in classifying patients with depressive disorder, as evidenced by the set of interpretable experiments, and can be an effective solution for disease control.

Discovering pathway biomarkers of hepatocellular carcinoma occurrence and development by dynamic network entropy analysis

OBJECTIVE

Gene expression profiling techniques measure the transcription of thousands of genes in a parallel manner. With more and more hepatocellular carcinoma (HCC) transcriptomic data becoming available, the high-throughput data provides an unprecedented opportunity to discover HCC diagnostic biomarkers. In this work, we propose a bioinformatics method based on dynamic network entropy analysis, called DNEA, to identify potential pathway biomarkers for HCC occurrence and development by integrating transcriptome and interactome.

METHODS

We firstly collect the pathways documented in different knowledge-bases and then impose the genome-wide human transcriptomic data of multistage cancerous tissues during the development and progression of HCC. After linking the gene sets of pathways into individual connected networks, we map the corresponding gene expression information onto these pathways. The dynamic network entropy of individual pathways is calculated to evaluate its activities and dysfunctionalities during the disease occurrence and development. We use the overall significant difference in the entropic dynamics during the time course to prioritize distinctive pathways during disease progression. Then machine learning classification methods are employed to screen out pathway biomarkers with the classification ability to distinguish different-stage samples of HCC progression.

RESULTS

Pathway biomarkers discovered based on DNEA demonstrate good classification performance in measuring HCC progression. The classification accuracy is as follows: DNA replication pathway (mean AUC= 0.82, 20 genes) from KEGG, FMLP pathway (mean AUC=0.84, 14 genes) from BioCarta, and downstream signaling of activated FGFR pathway (mean AUC =0.80, 15 genes) from Reactome. At the same time, previous studies have shown that these genes and pathways screened are closely related to the occurrence and development of HCC in terms of oncogenesis dysfunctions.

CONCLUSIONS

Our method for cancer biomarker discovery based on dynamic network entropy analysis is effective and efficient in identifying pathway biomarkers related to the progression of complex diseases.

Machine-learning predicts time-series prognosis factors in metastatic prostate cancer patients treated with androgen deprivation therapy

Machine learning technology is expected to support diagnosis and prognosis prediction in medicine. We used machine learning to construct a new prognostic prediction model for prostate cancer patients based on longitudinal data obtained from age at diagnosis, peripheral blood and urine tests of 340 prostate cancer patients. Random survival forest (RSF) and survival tree were used for machine learning. In the time-series prognostic prediction model for metastatic prostate cancer patients, the RSF model showed better prediction accuracy than the conventional Cox proportional hazards model for almost all time periods of progression-free survival (PFS), overall survival (OS) and cancer-specific survival (CSS). Based on the RSF model, we created a clinically applicable prognostic prediction model using survival trees for OS and CSS by combining the values of lactate dehydrogenase (LDH) before starting treatment and alkaline phosphatase (ALP) at 120 days after treatment. Machine learning provides useful information for predicting the prognosis of metastatic prostate cancer prior to treatment intervention by considering the nonlinear and combined impacts of multiple features. The addition of data after the start of treatment would allow for more precise prognostic risk assessment of patients and would be beneficial for subsequent treatment selection.

Prediction of hearing preservation after acoustic neuroma surgery based on SMOTE-XGBoost

Prior to the surgical removal of an acoustic neuroma, the majority of patients anticipate that their hearing will be preserved to the greatest possible extent following surgery. This paper proposes a postoperative hearing preservation prediction model for the characteristics of class-imbalanced hospital real data based on the extreme gradient boost tree (XGBoost). In order to eliminate sample imbalance, the synthetic minority oversampling technique (SMOTE) is applied to increase the number of underclass samples in the data. Multiple machine learning models are also used for the accurate prediction of surgical hearing preservation in acoustic neuroma patients. In comparison to research results from existing literature, the experimental results found the model proposed in this paper to be superior. In summary, the method this paper proposes can make a significant contribution to the development of personalized preoperative diagnosis and treatment plans for patients, leading to effective judgment for the hearing retention of patients with acoustic neuroma following surgery, a simplified long medical treatment process and saved medical resources.

Homologous recombination deficiency in triple-negative breast cancer: Multi-scale transcriptomics reveals distinct tumor microenvironments and limitations in predicting immunotherapy response

BACKGROUND

Triple-negative breast cancer (TNBC) is the most aggressive subtype of breast cancer and has the highest proportion of homologous recombination deficiency (HRD). HRD has been considered a biomarker of response to immune checkpoint inhibitors (ICIs), but the reality is more complicated. A comprehensive comparison of the tumor microenvironment (TME) in HRD and non-HRD TNBC samples may be helpful.

METHODS

Datasets from single-cell, spatial, and bulk RNA-sequencing were collected to explore the role of HRD in the development of TME at multiple scales. Based on the findings in the TME, machine learning algorithms were used to construct a response prediction model in eleven ICI therapy cohorts.

RESULTS

A more exhausted phenotype of T cells and a more tolerogenic phenotype of dendritic cells were found in the non-HRD group. HRD reprograms the predominant phenotype of cancer-associated fibroblasts (CAFs) from myofibroblastic CAFs to inflammatory-like CAFs. As interactions between myofibroblastic CAFs and other cells, DPP4-chemokines associated with reduced immune cell recruitment were unique in the non-HRD group. The prediction model based on DPP4-related genes had acceptable performance in predicting response, prognosis, and immune cell content. Higher HRD scores in bulk RNA-sequencing samples indicated more activated immune cell function, but not higher immune cell content, which may be affected by factors such as antigen-presenting capacity.

CONCLUSIONS

Based on multi-scale transcriptomics, our findings comprehensively reveal differences in the TME between HRD and non-HRD samples. Combining HRD with the prediction model or other methods for assessing immune cell content, may better predict response to ICIs in TNBC.

Biochemical reaction network topology defines dose-dependent Drug-Drug interactions

Drug combination therapy is a promising strategy to enhance the desired therapeutic effect, while reducing side effects. High-throughput pairwise drug combination screening is a commonly used method for discovering favorable drug interactions, but is time-consuming and costly. Here, we investigate the use of reaction network topology-guided design of combination therapy as a predictive in silico drug-drug interaction screening approach. We focused on three-node enzymatic networks, with general Michaelis-Menten kinetics. The results revealed that drug-drug interactions critically depend on the choice of target arrangement in a given topology, the nature of the drug, and the desired level of change in the network output. The results showed a negative correlation between antagonistic interactions and the dosage of drugs. Overall, the negative feedback loops showed the highest synergistic interactions (the lowest average combination index) and, intriguingly, required the highest drug doses compared to other topologies under the same condition.

Automatic Differentiation for Inverse Problems in X-ray Imaging and Microscopy

Computational techniques allow breaking the limits of traditional imaging methods, such as time restrictions, resolution, and optics flaws. While simple computational methods can be enough for highly controlled microscope setups or just for previews, an increased level of complexity is instead required for advanced setups, acquisition modalities or where uncertainty is high; the need for complex computational methods clashes with rapid design and execution. In all these cases, Automatic Differentiation, one of the subtopics of Artificial Intelligence, may offer a functional solution, but only if a GPU implementation is available. In this paper, we show how a framework built to solve just one optimisation problem can be employed for many different X-ray imaging inverse problems.

Artificial Intelligence and Cancer Control: Toward Prioritizing Justice, Equity, Diversity, and Inclusion (JEDI) in Emerging Decision Support Technologies

PURPOSE FOR REVIEW

This perspective piece has two goals: first, to describe issues related to artificial intelligence-based applications for cancer control as they may impact health inequities or disparities; and second, to report on a review of systematic reviews and meta-analyses of artificial intelligence-based tools for cancer control to ascertain the extent to which discussions of justice, equity, diversity, inclusion, or health disparities manifest in syntheses of the field's best evidence.

RECENT FINDINGS

We found that, while a significant proportion of existing syntheses of research on AI-based tools in cancer control use formal bias assessment tools, the fairness or equitability of models is not yet systematically analyzable across studies. Issues related to real-world use of AI-based tools for cancer control, such as workflow considerations, measures of usability and acceptance, or tool architecture, are more visible in the literature, but still addressed only in a minority of reviews. Artificial intelligence is poised to bring significant benefits to a wide range of applications in cancer control, but more thorough and standardized evaluations and reporting of model fairness are required to build the evidence base for AI-based tool design for cancer and to ensure that these emerging technologies promote equitable healthcare.

Nucleotide-Binding Oligomerization Domain (NOD)-Like Receptor Subfamily C (NLRC) as a Prognostic Biomarker for Glioblastoma Multiforme Linked to Tumor Microenvironment: A Bioinformatics, Immunohistochemistry, and Machine Learning-Based Study

Purpose

Glioblastoma multiforme (GBM) remains the deadliest primary brain tumor. We aimed to illuminate the role of nucleotide-binding oligomerization domain (NOD)-like receptor subfamily C (NLRC) in GBM.

Patients and Methods

Based on public database data (mainly The Cancer Genome Atlas [TCGA]), we performed bioinformatics analysis to visually evaluate the role and mechanism of NLRCs in GBM. Then, we validated our findings in a glioma tissue microarray (TMA) by immunohistochemistry (IHC), and the prognostic value of NOD1 was assessed via random forest (RF) models.

Results

In GBM tissues, the expression of NLRC members was significantly increased, which was related to the low survival rate of GBM. Additionally, Cox regression analysis revealed that the expression of NOD1 (among NLRCs) served as an independent prognostic marker. A nomogram based on multivariate analysis proved the effective predictive performance of NOD1 in GBM. Enrichment analysis showed that high expression of NOD1 could regulate extracellular structure, cell adhesion, and immune response to promote tumor progression. Then, immune infiltration analysis showed that NOD1 overexpression correlated with an enhanced immune response. Then, in a glioma TMA, the results of IHC revealed that the increase in NOD1 expression indicated high recurrence and poor prognosis of human glioma. Furthermore, the expression level of NOD1 showed good prognostic value in the TMA cohort via RF.

Conclusion

The value of NOD1 as a biomarker for GBM was demonstrated. The possible mechanisms may lie in the regulatory role of NLRC-related pathways in the tumor microenvironment.

EMSI-BERT: Asymmetrical Entity-Mask Strategy and Symbol-Insert Structure for Drug–Drug Interaction Extraction Based on BERT

Drug-drug interaction (DDI) extraction has seen growing usage of deep models, but their effectiveness has been restrained by limited domain-labeled data, a weak representation of co-occurring entities, and poor adaptation of downstream tasks. This paper proposes a novel EMSI-BERT method for drug–drug interaction extraction based on an asymmetrical Entity-Mask strategy and a Symbol-Insert structure. Firstly, the EMSI-BERT method utilizes the asymmetrical Entity-Mask strategy to address the weak representation of co-occurring entity information using the drug entity dictionary in the pre-training BERT task. Secondly, the EMSI-BERT method incorporates four symbols to distinguish different entity combinations of the same input sequence and utilizes the Symbol-Insert structure to address the week adaptation of downstream tasks in the fine-tuning stage of DDI classification. The experimental results showed that EMSI-BERT for DDI extraction achieved a 0.82 F1-score on DDI-Extraction 2013, and it improved the performances of the multi-classification task of DDI extraction and the two-classification task of DDI detection. Compared with baseline Basic-BERT, the proposed pre-training BERT with the asymmetrical Entity-Mask strategy could obtain better effects in downstream tasks and effectively limit “Other” samples’ effects. The model visualization results illustrated that EMSI-BERT could extract semantic information at different levels and granularities in a continuous space.

Explainable artificial intelligence as a reliable annotator of archaeal promoter regions

Archaea are a vast and unexplored cellular domain that thrive in a high diversity of environments, having central roles in processes mediating global carbon and nutrient fluxes. For these organisms to balance their metabolism, the appropriate regulation of their gene expression is essential. A key momentum in regulating genes responsible for the life maintenance of archaea is when transcription factor proteins bind to the promoter element. This DNA segment is conserved, which enables its exploration by machine learning techniques. Here, we trained and tested a support vector machine with 3935 known archaeal promoter sequences. All promoter sequences were coded into DNA Duplex Stability. After, we performed a model interpretation task to map the decision pattern of the classification procedure. We also used a dataset of known-promoter sequences for validation. Our results showed that an AT rich region around position - 27 upstream (relative to the start TSS) is the most conserved in the analyzed organisms. In addition, we were able to identify the BRE element (- 33), the PPE (at - 10) and a position at + 3, that provides a more understandable picture of how promoters are organized in all the archaeal organisms. Finally, we used the interpreted model to identify potential promoter sequences of 135 unannotated organisms, delivering regulatory regions annotation of archaea in a scale never accomplished before ( https://pcyt.unam.mx/gene-regulation/ ). We consider that this approach will be useful to understand how gene regulation is achieved in other organisms apart from the already established transcription factor binding sites.

Artificial Intelligence and Data Mining for the Pharmacovigilance of Drug-Drug Interactions

Despite increasing mechanistic understanding, undetected and underrecognized drug-drug interactions (DDIs) persist. This elusiveness relates to an interwoven complexity of increasing polypharmacy, multiplex mechanistic pathways, and human biological individuality. This persistent elusiveness motivates development of artificial intelligence (AI)-based approaches to enhancing DDI detection and prediction capabilities. The literature is vast and roughly divided into "prediction" and "detection." The former relatively emphasizes biological and chemical knowledge bases, drug development, new drugs, and beneficial interactions, whereas the latter utilizes more traditional sources such as spontaneous reports, claims data, and electronic health records to detect novel adverse DDIs with authorized drugs. However, it is not a bright line, either nominally or in practice, and both are in scope for pharmacovigilance supporting signal detection but also signal refinement and evaluation, by providing data-based mechanistic arguments for/against DDI signals. The wide array of intricate and elegant methods has expanded the pharmacovigilance tool kit. How much they add to real prospective pharmacovigilance, reduce the public health impact of DDIs, and at what cost in terms of false alarms amplified by automation bias and its sequelae are open questions. (Clin Ther. 2023;45:XXX-XXX) © 2023 Elsevier HS Journals, Inc.

Big data- and artificial intelligence-based hot-spot analysis of COVID-19: Gauteng, South Africa, as a case study

The coronavirus disease 2019 (COVID-19) has developed into a pandemic. Data-driven techniques can be used to inform and guide public health decision- and policy-makers. In generalizing the spread of a virus over a large area, such as a province, it must be assumed that the transmission occurs as a stochastic process. It is therefore very difficult for policy and decision makers to understand and visualize the location specific dynamics of the virus on a more granular level. A primary concern is exposing local virus hot-spots, in order to inform and implement non-pharmaceutical interventions. A hot-spot is defined as an area experiencing exponential growth relative to the generalised growth of the pandemic. This paper uses the first and second waves of the COVID-19 epidemic in Gauteng Province, South Africa, as a case study. The study aims provide a data-driven methodology and comprehensive case study to expose location specific virus dynamics within a given area. The methodology uses an unsupervised Gaussian Mixture model to cluster cases at a desired granularity. This is combined with an epidemiological analysis to quantify each cluster's severity, progression and whether it can be defined as a hot-spot.

Developing a machine learning model to predict patient need for computed tomography imaging in the emergency department

Overcrowding is a well-known problem in hospitals and emergency departments (ED) that can negatively impact patients and staff. This study aims to present a machine learning model to detect a patient's need for a Computed Tomography (CT) exam in the emergency department at the earliest possible time. The data for this work was collected from ED at Thunder Bay Regional Health Sciences Centre over one year (05/2016-05/2017) and contained administrative triage information. The target outcome was whether or not a patient required a CT exam. Multiple combinations of text embedding methods, machine learning algorithms, and data resampling methods were experimented with to find the optimal model for this task. The final model was trained with 81, 118 visits and tested on a hold-out test set with a size of 9, 013 visits. The best model achieved a ROC AUC score of 0.86 and had a sensitivity of 87.3% and specificity of 70.9%. The most important factors that led to a CT scan order were found to be chief complaint, treatment area, and triage acuity. The proposed model was able to successfully identify patients needing a CT using administrative triage data that is available at the initial stage of a patient's arrival. By determining that a CT scan is needed early in the patient's visit, the ED can allocate resources to ensure these investigations are completed quickly and patient flow is maintained to reduce overcrowding.

Machine learning model identifies aggressive acute pancreatitis within 48 h of admission: a large retrospective study

BACKGROUND

Acute pancreatitis (AP) with critical illness is linked to increased morbidity and mortality. Current risk scores to identify high-risk AP patients have certain limitations.

OBJECTIVE

To develop and validate a machine learning tool within 48 h after admission for predicting which patients with AP will develop critical illness based on ubiquitously available clinical, laboratory, and radiologic variables.

METHODS

5460 AP patients were enrolled. Clinical, laboratory, and imaging variables were collected within 48 h after hospital admission. Least Absolute Shrinkage Selection Operator with bootstrap method was employed to select the most informative variables. Five different machine learning models were constructed to predictive likelihood of critical illness, and the optimal model (APCU) was selected. External cohort was used to validate APCU. APCU and other risk scores were compared using multivariate analysis. Models were evaluated by area under the curve (AUC). The decision curve analysis was employed to evaluate the standardized net benefit.

RESULTS

Xgboost was constructed and selected as APCU, involving age, comorbid disease, mental status, pulmonary infiltrates, procalcitonin (PCT), neutrophil percentage (Neu%), ALT/AST, ratio of albumin and globulin, cholinesterase, Urea, Glu, AST and serum total cholesterol. The APCU performed excellently in discriminating AP risk in internal cohort (AUC = 0.95) and external cohort (AUC = 0.873). The APCU was significant for biliogenic AP (OR = 4.25 [2.08-8.72], P < 0.001), alcoholic AP (OR = 3.60 [1.67-7.72], P = 0.001), hyperlipidemic AP (OR = 2.63 [1.28-5.37], P = 0.008) and tumor AP (OR = 4.57 [2.14-9.72], P < 0.001). APCU yielded the highest clinical net benefit, comparatively.

CONCLUSION

Machine learning tool based on ubiquitously available clinical variables accurately predicts the development of AP, optimizing the management of AP.