Realization Of Precision Medicine Research Articles

A High-Throughput and Logarithm-Serial-Dilution Microfluidic Chip for Combinational Drug Screening on Tumor Organoids.

Tumor organoids are biological models for studying precision medicine. Microfluidic technology offers significant benefits for high throughput drug screening using tumor organoids. However, the range of concentrations achievable with traditional linear gradient generators in microfluidics is restricted, generating logarithmic drug concentration gradients by adjusting the channel ratio in the chip is confined to single-drug dilution chips, significantly restricting the application of microfluidics in drug screening. Here, we presented a microfluidic chip featuring continuous dilution capabilities, which generates logarithmic stepwise drug concentration gradients. We have devised a "mathematical-circuit-chip" model for designing such chips, and based on this model, we have developed and fabricated a device capable of providing 36 distinct drug concentration conditions for two types of drugs. The chip is composed of two structurally identical yet orthogonally arranged layers, each containing a dilution network capable of forming a 5-fold gradient and a tumor organoid culture module. Drug and culture medium delivery to the open culture chamber array is driven by syringe pumps. We have conducted drug screening experiments on patient-derived tumor organoids. This device facilitates high-throughput drug screening for patient-derived organoids, representing a significant stride toward the realization of precision medicine.

Discovering explainable biomarkers for breast cancer anti-PD1 response via network Shapley value analysis

Background and objectiveImmunotherapy holds promise in enhancing pathological complete response rates in breast cancer, albeit confined to a select cohort of patients. Consequently, pinpointing factors predictive of treatment responsiveness is of paramount importance. Gene expression and regulation, inherently operating within intricate networks, constitute fundamental molecular machinery for cellular processes and often serve as robust biomarkers. Nevertheless, contemporary feature selection approaches grapple with two key challenges: opacity in modeling and scarcity in accounting for gene-gene interactions MethodsTo address these limitations, we devise a novel feature selection methodology grounded in cooperative game theory, harmoniously integrating with sophisticated machine learning models. This approach identifies interconnected gene regulatory network biomarker modules with priori genetic linkage architecture. Specifically, we leverage Shapley values on network to quantify feature importance, while strategically constraining their integration based on network expansion principles and nodal adjacency, thereby fostering enhanced interpretability in feature selection. We apply our methods to a publicly available single-cell RNA sequencing dataset of breast cancer immunotherapy responses, using the identified feature gene set as biomarkers. Functional enrichment analysis with independent validations further illustrates their effective predictive performance ResultsWe demonstrate the sophistication and excellence of the proposed method in data with network structure. It unveiled a cohesive biomarker module encompassing 27 genes for immunotherapy response. Notably, this module proves adept at precisely predicting anti-PD1 therapeutic outcomes in breast cancer patients with classification accuracy of 0.905 and AUC value of 0.971, underscoring its unique capacity to illuminate gene functionalities ConclusionThe proposed method is effective for identifying network module biomarkers, and the detected anti-PD1 response biomarkers can enrich our understanding of the underlying physiological mechanisms of immunotherapy, which have a promising application for realizing precision medicine.

MMFP-Tableau: enabling precision mitochondrial medicine through integration, visualization, and analytics of clinical and research health system electronic data.

To describe a novel data integration workflow developed to automate clinical and research electronic health system data integration and harmonization from siloed sources for centralized access, visualization and analysis by clinicians and researchers in an end user-friendly customized analytic platform. A centralized, semi-automated framework provides data provenance and user access to integrated data sources. Data models are implemented leveraging a centralized server (Alteryx) for high-level analytics including scheduling, integration, and modeling. Data are then sent to a secure Tableau instance for end-user visualization and interaction, with minimal software development required. MMFP-Tableau, named for its origin in the Mitochondrial Medicine Frontier Program (MMFP) at the Children's Hospital of Philadelphia (CHOP), has advanced efforts to realize precision medicine by facilitating expert clinician and researcher end-user direct access to integrated, highly robust health system datasets. This scalable data solution enables translational researchers to link subject and cohort clinical and research parameters with research samples; enhances external biopharma collaborations for clinical trial design, subject recruitment, and data tracking; accelerates retrospective clinical cohort data analysis; and improves complex data visualization for clinicians and researchers. MMFP-Tableau promotes complex data integration, visibility, and advanced analytic capabilities to facilitate seamless multidisciplinary research, benefitting clinical care and research in rare disease patients and cohorts. The MMFP-Tableau data platform integrates multiple data sources across various siloed data platforms to transform complex data into readily accessible datasets. This approach represents a generalizable workflow concept readily adaptable to implement across diverse fields of medicine.

Advancements in Nanoparticle-Based Precision Drug Delivery: A Review of Development and Optimization for Targeted Therapy in Preclinical Models

In recent years, the widespread adoption of nanoparticles has expanded across a broad spectrum of clinical domains. These nanoparticles have been specifically engineered to address the limitations associated with conventional therapeutics and to navigate various biological barriers—ranging from systemic to cellular levels—that manifest heterogeneity across different patient populations and disease states. The advent of precision treatments, wherein interventions are tailored to individual patients, has contributed to mitigating this variability among patients. Nonetheless, the predominant focus in current nanoparticle research remains on enhancing the uniformity of delivery systems. The realization of precision medicine appears imminent as lipid-based, polymeric, and inorganic nanoparticles are increasingly crafted with heightened precision, facilitating more individualized approaches to medication delivery. In this review, we delve into the advanced designs of nanoparticles employed in both precision and generalized applications, offering insights into their potential to advance precision medicine. Our discussion centers on the innovations in nanoparticle design aimed at overcoming various delivery challenges, suggesting that ingenious nanoparticle engineering holds promise for enhancing performance across a broad spectrum of delivery applications and facilitating tailored designs for specific therapeutic targets, ultimately leading to improved patient outcomes.

Realizing precision medicine in chronic lymphocytic leukemia: Remaining challenges and potential opportunities.

Patients with chronic lymphocytic leukemia (CLL) exhibit diverse clinical outcomes. An expanding array of genetic tests is now employed to facilitate the identification of patients with high-risk disease and inform treatment decisions. These tests encompass molecular cytogenetic analysis, focusing on recurrent chromosomal alterations, particularly del(17p). Additionally, sequencing is utilized to identify TP53 mutations and to determine the somatic hypermutation status of the immunoglobulin heavy variable gene. Concurrently, a swift advancement of targeted treatment has led to the implementation of novel strategies for patients with CLL, including kinase and BCL2 inhibitors. This review explores both current and emerging diagnostic tests aimed at identifying high-risk patients who should benefit from targeted therapies. We outline existing treatment paradigms, emphasizing the importance of matching the right treatment to the right patient beyond genetic stratification, considering the crucial balance between safety and efficacy. We also take into consideration the practical and logistical issues when choosing a management strategy for each individual patient. Furthermore, we delve into the mechanisms underlying therapy resistance and stress the relevance of monitoring measurable residual disease to guide treatment decisions. Finally, we underscore the necessity of aggregating real-world data, adopting a global perspective, and ensuring patient engagement. Taken together, we argue that precision medicine is not the mere application of precision diagnostics and accessibility of precision therapies in CLL but encompasses various aspects of the patient journey (e.g., lifestyle exposures and comorbidities) and their preferences toward achieving true personalized medicine for patients with CLL.

Prediction of Lung Cancer Based on Catboost

This research aims to provide guidance for achieving precision medicine by accurately predicting the incidence of lung cancer. This paper used random forest screening to identify several variables that make significant influences on the lung cancer, and used Smote oversampling to address the issue of data imbalance. Finally, this paper used Catboost to construct a model to handle categorical features. Through analyzing experimental data, can found that among the selected variables, age has the greatest importance. Afterwards, validated the model and the training score was 0.9032, indicating good results. At the same time, establish a confusion matrix to prove once again that the model has good predictive performance. Finally, based on cross validation, both the optimal validation accuracy score and the accurate validation accuracy score were above 0.9, indicating that the model performance is excellent. The innovation point of this paper is to realize precision medicine through high accuracy prediction of lung cancer.

Automatic Measurement of Loop Gain Based on Pseudorandom Binary Sequence CO2 Stimulation.

Measurement of loop gain in patients with obstructive sleep apnea (OSA) is of great importance for determining the cause of OSA and realizing precision medicine. In this study, automatic measurement of loop gain is carried out during wakefulness based on the pseudorandom binary sequence (PRBS) CO2 stimulation method. A respiratory data acquisition platform is designed and constructed to automate the processes of gas configuration, ventilatory stimulation with CO2 and data acquisition. The respiratory data are substituted into the model of the ventilatory control system for parameter identification, and the loop gain values are calculated and then compared with the apnea-hypopnea index (AHI) measured in a hypoxia chamber. A positive correlation is found between loop gain and AHI measured in the hypoxia chamber, with the linear correlation coefficient of approximately 0.65. The feasibility of automatic measurement of loop gain using the respiratory data acquisition platform based on the PRBS CO2 stimulation method is validated, and the measured loop gain values can be used to assess the stability of ventilatory control. This study provides an automated, rapid, and instrumented solution for loop gain measurement, laying the foundation for wide-scale clinical application of the PRBS CO2 stimulation method.

Best practices for multimodal clinical data management and integration: An atopic dermatitis research case

BackgroundIn clinical research on multifactorial diseases such as atopic dermatitis, data-driven medical research has become more widely used as means to clarify diverse pathological conditions and to realize precision medicine. However, modern clinical data, characterized as large-scale, multimodal, and multi-center, causes difficulties in data integration and management, which limits productivity in clinical data science. MethodsWe designed a generic data management flow to collect, cleanse, and integrate data to handle different types of data generated at multiple institutions by 10 types of clinical studies. We developed MeDIA (Medical Data Integration Assistant), a software to browse the data in an integrated manner and extract subsets for analysis. ResultsMeDIA integrates and visualizes data and information on research participants obtained from multiple studies. It then provides a sophisticated interface that supports data management and helps data scientists retrieve the data sets they need. Furthermore, the system promotes the use of unified terms such as identifiers or sampling dates to reduce the cost of pre-processing by data analysts. We also propose best practices in clinical data management flow, which we learned from the development and implementation of MeDIA. ConclusionsThe MeDIA system solves the problem of multimodal clinical data integration, from complex text data such as medical records to big data such as omics data from a large number of patients. The system and the proposed best practices can be applied not only to allergic diseases but also to other diseases to promote data-driven medical research.

3D hepatic organoid production from human pluripotent stem cells

Hepatic organoids might provide a golden opportunity for realizing precision medicine in various hepatic diseases. Previously described hepatic organoid protocols from pluripotent stem cells rely on complicated multiple differentiation steps consisting of both 2D and 3D differentiation procedures. Therefore, the spontaneous formation of hepatic organoids from 2D monolayer culture is associated with a low-throughput production, which might hinder the standardization of hepatic organoid production and hamper the translation of this technology to the clinical or industrial setting. Here we describe the stepwise and fully 3D production of hepatic organoids from human pluripotent stem cells. We optimized every differentiation step by screening for optimal concentrations and timing of differentiation signals in each differentiation step. Hepatic organoids are stably expandable without losing their hepatic functionality. Moreover, upon treatment of drugs with known hepatotoxicity, we found hepatic organoids are more sensitive to drug-induced hepatotoxicity compared with 2D hepatocytes differentiated from PSCs, making them highly suitable for in vitro toxicity screening of drug candidates. The standardized fully 3D protocol described in the current study for producing functional hepatic organoids might serve as a novel platform for the industrial and clinical translation of hepatic organoid technology.

Abstract TMP72: Multimodal Prediction Of Stroke Severity

Introduction: Predicting individual outcomes post-stroke with the highest possible accuracy is a crucial steppingstone in the realization of precision medicine. We here evaluated various types of lesion information in their capacity to predict stroke severity in a large cohort of patients with acute ischemic stroke. Methods: A total of 1,075 patients of the MRI-GENIE study [age: 64.2(14.7), 38% women] contributed to analyses (N=792 as train and N=283 as test sample). We employed ridge regression with hyperparameter optimization and nested five-fold cross-validation to predict acute NIHSS-based stroke severity. Our baseline model considered DWI lesion volume. Further models tested structural lesion location, indirect structural and functional lesion connectivity, individually and combined with lesion volume. Data was preprocessed by principal component analysis (PCA), retaining 95% of the variance of the original data. Model performance was compared in terms of explained variance (R-squared) and 95% confidence intervals in the outer loop of the training set and the test set. Results: Structural lesion connectivity enabled the highest prediction performance in the test set ( Figure 1 ). Prediction performance did not change notably with inclusion of lesion volume information. This contrasted with the combination of functional lesion connectivity and lesion volume that achieved a substantially higher prediction performance compared to functional lesion connectivity or lesion volume in isolation. Lesion location resulted in a prediction performance that was in the range of lesion volume alone and significantly lower than the ones of structural or functional connectivity with lesion volume. Conclusions: Structural connectivity facilitated the most convincing prediction of stroke severity. The capacity of functional lesion connectivity was substantially improved by the inclusion of lesion volume, indicating both represent complementary information.

Risk factor structure of heart failure in patients with cancer after treatment with anticancer agents’ assessment by big data from a Japanese electronic health record

As the prognosis of cancer patients has been improved, comorbidity of heart failure (HF) in cancer survivors is a serious concern, especially in the aged population. This study aimed to examine the risk factors of HF development after treatment by anticancer agents, using a machine learning-based analysis of a massive dataset obtained from the electronic health record (EHR) in Japan. This retrospective, cohort study, using a dataset from 2008 to 2017 in the Diagnosis Procedure Combination (DPC) database in Japan, enrolled 140,327 patients. The structure of risk factors was determined using multivariable analysis and classification and regression tree (CART) algorithm for time-to-event data. The mean follow-up period was 1.55 years. The prevalence of HF after anticancer agent administration were 4.0%. HF was more prevalent in the older than the younger. As the presence of cardiovascular diseases and various risk factors predicted HF, CART analysis of the risk factors revealed that the risk factor structures complicatedly differed among different age groups. The highest risk combination was hypertension, diabetes mellitus, and atrial fibrillation in the group aged ≤ 64 years, and the presence of ischemic heart disease was a key in both groups aged 65–74 years and 75 ≤ years. The machine learning-based approach was able to develop complicated HF risk structures in cancer patients after anticancer agents in different age population, of which knowledge would be essential for realizing precision medicine to improve the prognosis of cancer patients.

Enabling Precision Medicine with CRISPR-Cas Genome Editing Technology: A Translational Perspective.

Genome editing technologies, particularly CRISPR-Cas (clustered regularly interspaced short palindromic repeats (CRISPR) associated nucleases), are redefining the boundaries of therapeutic gene therapy. CRISPR-Cas is a robust, straightforward, and programmable genome editing tool capable of mediating site-specific DNA modifications. The rapid advancements from discovery to clinical adaptation have expanded the therapeutic landscape to treat genetically defined diseases. Together with the technical developments in human DNA and RNA sequencing, CRISPR-directed gene therapy enables a new era to realize precision medicine where pathogenic mutations underlying monogenic disorders can potentially be corrected. Also, protective or therapeutic genomic alterations can be introduced as preventative or curative therapy. Despite its high therapeutic potential, CRISPR-Cas´ clinical translation is still in its infancy and is highly dependent on its efficiency, specificity in gene corrections, and cell-specific delivery. Therefore, this chapter focuses on the challenges and opportunities the CRISPR-Cas toolbox offers together with delivery vehicles to realize its use for therapeutic gene editing. Furthermore, we discuss the obstacles the CRISPR-Cas system faces for successful clinical translation and summarize its current clinical progress.

The joint detection of CEA and ctDNA in cerebrospinal fluid: an auxiliary tool for the diagnosis of leptomeningeal metastases in cancer.

Leptomeningeal metastases (LMs) are highly invasive which leads to poor prognosis, but the accurate diagnosis of LM is challenging. It is necessary to investigate more advanced diagnostic methods to realize precision medicine. The main purpose of this study was to select a more effective method for the auxiliary diagnosis of LM by comparing various detection methods. The secondary purpose was to explore the value of cerebrospinal fluid (CSF) tumor markers (TMs) and circulating tumor DNA (ctDNA) testing in guiding clinical treatment. TMs in serum and CSF of patients were detected by chemiluminescence. The ctDNA of CSF and plasma were detected by the next-generation sequencing (NGS) technology. In total, 54 tumor patients participated in this study, in which 39 with LM and 15 without LM (8 with parenchymal tumor and 7 without brain metastasis). The results showed that the sensitivity and accuracy of CSF cytology isolated during the first lumbar puncture were 0.31 (95% CI 0.17-0.48) and 0.50 (95% CI 0.36-0.64), respectively. The sensitivity and accuracy of CSF_CEA were 0.71 (95% CI 0.54-0.85) and 0.78 (95% CI 0.64-0.89), which were better than those of CSF_NSE and CSF_CFRA-211. The sensitivity and accuracy of CSF_ctDNA were 0.92 (95% CI 0.79-0.98) and 0.91 (95% CI 0.80-0.97), significantly higher than that of CSF cytology (P < 0.01). The sensitivity and accuracy of CSF_CEA combined with CSF_ctDNA were 0.97 (95% CI, 0.87-1.00) and 0.94 (95% CI 0.85-0.99), which were significantly higher than the traditional methods CSF cytology (P < 0.01). For LM patients with hydrocephalus, the sensitivity of CSF ctDNA even achieved 100% (14/14). CSF_CEA combined with CSF_ctDNA could be used to accurately distinguish patients with LM from those with no brain metastasis and from those with parenchymal tumors. CSF_ctDNA testing reveals a unique mutation profile for patients with LM. Dynamic detection of CSF TM and ctDNA can better predict the efficacy and reveal the cause of drug resistance to guide subsequent treatment. Clinical trial registration number: NCT03029065.

Multidisciplinary Effort to Drive Precision-Medicine for the Future.

In the past one or two decades, countries across the world have successively implemented different precision medicine (PM) programs, and also cooperated to implement international PM programs. We are now in the era of PM. Singapore's National Precision Medicine (NPM) program, initiated in 2017, is now entering its second phase to generate a large genomic database for Asians. The National University of Singapore (NUS) also launched its own PM translational research program (TRP) in 2021, aimed at consolidating multidisciplinary expertise within the Yong Loo Lin School of Medicine to develop collaborative projects that can help to identify and validate novel therapeutic targets for the realization of PM. To achieve this, appropriate data collection, data processing, and results interpretation must be taken into consideration. There may be some difficulties during these processes, but with the improvement of relevant rules and the continuous development of omics-based technologies, we will be able to solve these problems, eventually achieving precise prediction, diagnosis, treatment, or even prevention of diseases.

Predictive and Prognostic Role of Peripheral Blood T-Cell Subsets in Triple-Negative Breast Cancer.

BackgroundTriple-negative breast cancer (TNBC), as a highly aggressive and metastatic tumor, can still not contain the medical needs. It has become an urgent problem to develop prognostic markers further and realize precision medicine. The predictive and prognostic significance of peripheral blood lymphocytes, as well as the clinicopathological factors affecting them, were explored in the present study.MethodsThe clinicopathological data of 278 patients with TNBC were collected and analyzed retrospectively. Peripheral blood lymphocytes (pBL) and blood routine indexes before treatment were quantified by flow cytometry analysis. Progression-free survival (PFS) and overall survival (OS) were analyzed by the Kaplan-Meier curve and Cox hazard proportion regression model. The associations between hematologic parameters and treatment response and clinicopathological characteristics were estimated by the Mann-Whitney test and Spearman test.ResultsCompared with all blood routine indexes, only a significant correlation between better treatment efficacy and higher peripheral CD4 +/CD8 + ratio of TNBC patients was observed (P=0.059), particularly those treated with chemotherapy combined with immune checkpoint inhibitors (P=0.048). Among the pBL subsets, CD4 + T lymphocyte was the only independent factor that could predict the prognosis of metastatic TNBC. Patients presenting higher values of peripheral CD4 + T lymphocytes showed longer PFS (median PFS 9 months vs. 5 months; HR =0.65, 95%CI: 0.440-0.973, P = 0.032) and OS (median OS 31 months vs. 16 months; HR=0 .63, 95%CI: 0.417-0.940, P< 0.01). Especially CD4+ was found predictive for prognosis in TNBC patients who received chemotherapy (P<0.05). Finally, the older age, higher clinical stage, and more advanced treatment lines were related to the lower level of CD4 +. The older age and having received neoadjuvant therapy were related to the lower CD4 +/CD8 + ratio (P<0.05).ConclusionThe baseline CD4+/CD8+ cell ratio in peripheral blood is associated with therapeutic response, especially for chemotherapy combined with immunotherapy. Peripheral CD4+ cells can steadily predict all clinical outcomes for patients with mTNBC, and this clinical prognosis prediction is significantly related to chemotherapy. Peripheral CD4+ and CD4+/CD8+ are both closely associated with clinicopathological parameters.

Biomarkers of acute kidney injury after pediatric cardiac surgery: a meta-analysis of diagnostic test accuracy.

A variety of biomarkers have been proposed as predictors of cardiac surgery-associated AKI in children, of which uNGAL was the most prominent with excellent diagnostic qualities. However, more consolidatory evidence will be required before these novel biomarkers may eventually help realize precision medicine in AKI management. • Acute kidney injury (AKI) occurs in about 30-60% of children undergoing cardiac surgery and is associated with increased in-hospital mortality and adverse short-term outcomes.However, in current clinical practice, AKI definitions and detection often rely on changes in serum creatinine and urine output, which are late and insensitive markers of kidney injury. • Although various novel biomarkers have been studied for the diagnosis of AKI in children after cardiac surgery, it remains unclear how these compare to one another in terms of diagnostic accuracy. • Pooled analyses suggest that for the diagnosis of AKI in children who underwent cardiac surgery, NGAL is the most accurate among the most frequently studied biomarkers. • A number of other promising biomarkers have been reported, although they will require further research into their diagnostic accuracy and clinical applicability.

Precision medicine in stroke: towards personalized outcome predictions using artificial intelligence.

Stroke ranks among the leading causes for morbidity and mortality worldwide. New and continuously improving treatment options such as thrombolysis and thrombectomy have revolutionized acute stroke treatment in recent years. Following modern rhythms, the next revolution might well be the strategic use of the steadily increasing amounts of patient-related data for generating models enabling individualized outcome predictions. Milestones have already been achieved in several health care domains, as big data and artificial intelligence have entered everyday life.The aim of this review is to synoptically illustrate and discuss how artificial intelligence approaches may help to compute single-patient predictions in stroke outcome research in the acute, subacute and chronic stage. We will present approaches considering demographic, clinical and electrophysiological data, as well as data originating from various imaging modalities and combinations thereof. We will outline their advantages, disadvantages, their potential pitfalls and the promises they hold with a special focus on a clinical audience. Throughout the review we will highlight methodological aspects of novel machine-learning approaches as they are particularly crucial to realize precision medicine. We will finally provide an outlook on how artificial intelligence approaches might contribute to enhancing favourable outcomes after stroke.

Getting personal about skin: Realizing precision medicine in dermatology

AbstractWith the massive influx of data from “omics” methods combined with sophisticated informatics platforms, personalized medicine, defined as the tailoring of medical decision‐making and/or treatments to an individual patient, has led to the promise of better and more efficient patient care. As a concept, personalized medicine could fundamentally alter how clinicians think about the diagnosis and treatment of disease. The introduction of a precision medicine tool has benefits for multiple stakeholders; patients benefit by having improved outcomes, physicians can make confident decisions in managing their patients, and the healthcare system saves significant dollars through the reduction of wasted spend. This paradigm has been particularly true in oncology where biomarker‐driven treatment paradigms have become increasingly common. In this review, personalized medicine and biomarkers in the field of dermatology are reviewed. Specifically, three disease states are discussed including psoriasis, atopic dermatitis, and melanoma. For each disease state, biomarkers and personalized medicine tests are reviewed focusing on diagnosis and treatment for these conditions. The ethics, implementation, advantages, and disadvantages are discussed for the adaptation of personalized medicine in clinical dermatology. The advent of personalized medicine in dermatology could lead to a paradigm shift in how patients are treated, with the resulting improved clinical outcomes leading to concomitant reductions in wasted healthcare expenditures.

Time-to-event prediction analysis of patients with chronic heart failure comorbid with atrial fibrillation: a LightGBM model

BackgroundChronic heart failure (CHF) comorbid with atrial fibrillation (AF) is a serious threat to human health and has become a major clinical burden. This prospective cohort study was performed to design a risk stratification system based on the light gradient boosting machine (LightGBM) model to accurately predict the 1- to 3-year all-cause mortality of patients with CHF comorbid with AF.MethodsElectronic medical records of hospitalized patients with CHF comorbid with AF from January 2014 to April 2019 were collected. The data set was randomly divided into a training set and test set at a 3:1 ratio. In the training set, the synthetic minority over-sampling technique (SMOTE) algorithm and fivefold cross validation were used for LightGBM model training, and the model performance was performed on the test set and compared using the logistic regression method. The survival rate was presented on a Kaplan–Meier curve and compared by a log-rank test, and the hazard ratio was calculated by a Cox proportional hazard model.ResultsOf the included 1796 patients, the 1-, 2-, and 3-year cumulative mortality rates were 7.74%, 10.63%, and 12.43%, respectively. Compared with the logistic regression model, the LightGBM model showed better predictive performance, the area under the receiver operating characteristic curve for 1-, 2-, and 3-year all-cause mortality was 0.718 (95%CI, 0.710–0.727), 0.744(95%CI, 0.737–0.751), and 0.757 (95%CI, 0.751–0.763), respectively. The net reclassification index was 0.062 (95%CI, 0.044–0.079), 0.154 (95%CI, 0.138–0.172), and 0.148 (95%CI, 0.133–0.164), respectively. The differences between the two models were statistically significant (P < 0.05). Patients in the high-risk group had a significantly higher hazard of death than those in the low-risk group (hazard ratios: 12.68, 13.13, 14.82, P < 0.05).ConclusionRisk stratification based on the LightGBM model showed better discriminative ability than traditional model in predicting 1- to 3-year all-cause mortality of patients with CHF comorbid with AF. Individual patients’ prognosis could also be obtained, and the subgroup of patients with a higher risk of mortality could be identified. It can help clinicians identify and manage high- and low-risk patients and carry out more targeted intervention measures to realize precision medicine and the optimal allocation of health care resources.

High-resolution transcription factor binding sites prediction improved performance and interpretability by deep learning method.

Transcription factors (TFs) are essential proteins in regulating the spatiotemporal expression of genes. It is crucial to infer the potential transcription factor binding sites (TFBSs) with high resolution to promote biology and realize precision medicine. Recently, deep learning-based models have shown exemplary performance in the prediction of TFBSs at the base-pair level. However, the previous models fail to integrate nucleotide position information and semantic information without noisy responses. Thus, there is still room for improvement. Moreover, both the inner mechanism and prediction results of these models are challenging to interpret. To this end, the Deep Attentive Encoder-Decoder Neural Network (D-AEDNet) is developed to identify the location of TFs-DNA binding sites in DNA sequences. In particular, our model adopts Skip Architecture to leverage the nucleotide position information in the encoder and removes noisy responses in the information fusion process by Attention Gate. Simultaneously, the Transcription Factor Motif Discovery based on Sliding Window (TF-MoDSW), an approach to discover TFs-DNA binding motifs by utilizing the output of neural networks, is proposed to understand the biological meaning of the predicted result. On ChIP-exo datasets, experimental results show that D-AEDNet has better performance than competing methods. Besides, we authenticate that Attention Gate can improve the interpretability of our model by ways of visualization analysis. Furthermore, we confirm that ability of D-AEDNet to learn TFs-DNA binding motifs outperform the state-of-the-art methods and availability of TF-MoDSW to discover biological sequence motifs in TFs-DNA interaction by conducting experiment on ChIP-seq datasets.