Biologically informed deep neural network for prostate cancer discovery

Novel natural vector with asymmetric covariance for classifying biological sequences

The genome sequences of organisms form a large and complex landscape, presenting a significant challenge in bioinformatics: how to utilize mathematical tools to describe and analyze this space effectively. The ability to compare relationships between different organisms depends on creating a rational mapping rule that can uniformly encode genome sequences of varying lengths as vectors in a measurable space. This mapping would enable researchers to apply modern mathematical and machine learning techniques to otherwise challenging genomic comparisons. The natural vector method has been proposed as a concise and effective approach to accomplish this. However, its various iterations have certain limitations. In response, we carefully analyze the strengths and weaknesses of these natural vector methods and propose an improved version-an asymmetric covariance natural vector method (ACNV). This new method incorporates k-mer information alongside covariance computations with asymmetric properties between base positions. We tested ACNV on microbial genome sequence datasets, including bacterial, fungal, and viral sequences, evaluating its performance in terms of classification accuracy and convex hull separation. The results demonstrate that ACNV effectively captures sequence characteristics, showcasing its robust sequence representation capabilities and highlighting its elegant geometric properties.

DeepQA: A Unified Transcriptome-Based Aging Clock Using Deep Neural Networks

Understanding the complex biological process of aging is of great value, especially as it can help develop therapeutics to prolong healthy life. Predicting biological age from gene expression data has shown to be an effective means to quantify aging of a subject, and to identify molecular and cellular biomarkers of aging. A typical approach for estimating biological age, adopted by almost all existing aging clocks, is to train machine learning models only on healthy subjects, but to infer on both healthy and unhealthy subjects. However, the inherent bias in this approach results in inaccurate biological age as shown in this study. Moreover, almost all existing transcriptome-based aging clocks were built around an inefficient procedure of gene selection followed by conventional machine learning models such as elastic nets, linear discriminant analysis etc. To address these limitations, we proposed DeepQA, a unified aging clock based on mixture of experts. Unlike existing methods, DeepQA is equipped with a specially designed Hinge-Mean-Absolute-Error (Hinge-MAE) loss so that it can train on both healthy and unhealthy subjects of multiple cohorts to reduce the bias of inferring biological age of unhealthy subjects. Our experiments showed that DeepQA significantly outperformed existing methods for biological age estimation on both healthy and unhealthy subjects. In addition, our method avoids the inefficient exhaustive search of genes, and provides a novel means to identify genes activated in aging prediction, alternative to such as differential gene expression analysis.

Design of Multi-Cancer VOCs Profiling Platform via a Deep Learning-Assisted Sensing Library Screening Strategy

The efficiency of sensor arrays in parallel discrimination of multianalytes is fundamentally influenced by the quantity and performance of the sensor elements. The advent of combinational design has notably accelerated the generation of chemical libraries, offering numerous candidates for the development of robust sensor arrays. However, screening elements with superior cross-responsiveness remains challenging, impeding the development of high-performance sensor arrays. Herein, we propose a new deep learning-assisted, two-step screening strategy to identify the optimal combination of minimal sensor elements, using a designed volatile organic compounds (VOCs)-targeted sensor library. 400 sensing elements constructed by pairing 20 ionizable cationic elements and 20 anionic dyes in the sensor library were employed for various VOCs, generating plentiful color variation data. By employing a feedforward neural network─random forest-recursive feature elimination (FRR) algorithm, sensing elements were effectively screened, resulting in the rapidly producing 8-element and 10-element arrays for two VOC models, both achieving 100% discrimination accuracy. Furthermore, a smartphone-based point-of-care testing (POCT) platform achieved cancer discrimination in a simulated cancer VOC model, using image-based deep learning, demonstrating the rationality and practicality of deep learning in the assembly of sensor elements for parallel sensing platforms.

c-Triadem: A constrained, explainable deep learning model to identify novel biomarkers in Alzheimer's disease

Alzheimer's disease (AD) is a neurodegenerative disorder that requires early diagnosis for effective management. However, issues with currently available diagnostic biomarkers preclude early diagnosis, necessitating the development of alternative biomarkers and methods, such as blood-based diagnostics. We propose c-Triadem (constrained triple-input Alzheimer's disease model), a novel deep neural network to identify potential blood-based biomarkers for AD and predict mild cognitive impairment (MCI) and AD with high accuracy. The model utilizes genotyping data, gene expression data, and clinical information to predict the disease status of participants, i.e., cognitively normal (CN), MCI, or AD. The nodes of the neural network represent genes and their related pathways, and the edges represent known relationships among the genes and pathways. Simulated data validation further highlights the robustness of key features identified by SHapley Additive exPlanations (SHAP). We trained the model with blood genotyping data, microarray, and clinical features from the Alzheimer's Neuroimaging Disease Initiative (ADNI). We demonstrate that our model's performance is superior to previous models with an AUC of 97% and accuracy of 89%. We then identified the most influential genes and clinical features for prediction using SHapley Additive exPlanations (SHAP). Our SHAP analysis shows that CASP9, LCK, and SDC3 SNPs and PINK1, ATG5, and ubiquitin (UBB, UBC) expression have a higher impact on model performance. Our model has facilitated the identification of potential blood-based genetic markers of DNA damage response and mitophagy in affected regions of the brain. The model can be used for detection and biomarker identification in other related dementias.

New horizons at the interface of artificial intelligence and translational cancer research

Artificial intelligence (AI) is increasingly being utilized in cancer research as a computational strategy for analyzing multiomics datasets. Advances in single-cell and spatial profiling technologies have contributed significantly to our understanding of tumor biology, and AI methodologies are now being applied to accelerate translational efforts, including target discovery, biomarker identification, patient stratification, and therapeutic response prediction. Despite these advancements, the integration of AI into clinical workflows remains limited, presenting both challenges and opportunities. This review discusses AI applications in multiomics analysis and translational oncology, emphasizing their role in advancing biological discoveries and informing clinical decision-making. Key areas of focus include cellular heterogeneity, tumor microenvironment interactions, and AI-aided diagnostics. Challenges such as reproducibility, interpretability of AI models, and clinical integration are explored, with attention to strategies for addressing these hurdles. Together, these developments underscore the potential of AI and multiomics to enhance precision oncology and contribute to advancements in cancer care.

A critical assessment of artificial intelligence in magnetic resonance imaging of cancer

Given the enormous output and pace of development of artificial intelligence (AI) methods in medical imaging, it can be challenging to identify the true success stories to determine the state-of-the-art of the field. This report seeks to provide the magnetic resonance imaging (MRI) community with an initial guide into the major areas in which the methods of AI are contributing to MRI in oncology. After a general introduction to artificial intelligence, we proceed to discuss the successes and current limitations of AI in MRI when used for image acquisition, reconstruction, registration, and segmentation, as well as its utility for assisting in diagnostic and prognostic settings. Within each section, we attempt to present a balanced summary by first presenting common techniques, state of readiness, current clinical needs, and barriers to practical deployment in the clinical setting. We conclude by presenting areas in which new advances must be realized to address questions regarding generalizability, quality assurance and control, and uncertainty quantification when applying MRI to cancer to maintain patient safety and practical utility.

Soft Materials and Devices Enabling Sensorimotor Functions in Soft Robots

Sensorimotor functions, the seamless integration of sensing, decision-making, and actuation, are fundamental for robots to interact with their environments. Inspired by biological systems, the incorporation of soft materials and devices into robotics holds significant promise for enhancing these functions. However, current robotics systems often lack the autonomy and intelligence observed in nature due to limited sensorimotor integration, particularly in flexible sensing and actuation. As the field progresses toward soft, flexible, and stretchable materials, developing such materials and devices becomes increasingly critical for advanced robotics. Despite rapid advancements individually in soft materials and flexible devices, their combined applications to enable sensorimotor capabilities in robots are emerging. This review addresses this emerging field by providing a comprehensive overview of soft materials and devices that enable sensorimotor functions in robots. We delve into the latest development in soft sensing technologies, actuation mechanism, structural designs, and fabrication techniques. Additionally, we explore strategies for sensorimotor control, the integration of artificial intelligence (AI), and practical application across various domains such as healthcare, augmented and virtual reality, and exploration. By drawing parallels with biological systems, this review aims to guide future research and development in soft robots, ultimately enhancing the autonomy and adaptability of robots in unstructured environments.

Modeling microbiome-trait associations with taxonomy-adaptive neural networks

The human microbiome, a complex ecosystem of microorganisms inhabiting the body, plays a critical role in human health. Investigating its association with host traits is essential for understanding its impact on various diseases. Although shotgun metagenomic sequencing technologies have produced vast amounts of microbiome data, analyzing such data is highly challenging due to its sparsity, noisiness, and high feature dimensionality. Here, we develop MIOSTONE, an accurate and interpretable neural network model for microbiome-disease association that simulates a real taxonomy by encoding the relationships among microbial features. The taxonomy-encoding architecture provides a natural bridge from variations in microbial taxa abundance to variations in traits, encompassing increasingly coarse scales from species to domains. MIOSTONE has the ability to determine whether taxa within the corresponding taxonomic group provide a better explanation in a data-driven manner. MIOSTONE serves as an effective predictive model, as it not only accurately predicts microbiome-trait associations across extensive simulated and real datasets but also offers interpretability for scientific discovery. Both attributes are crucial for facilitating in silico investigations into the biological mechanisms underlying such associations among microbial taxa. Video Abstract.

Strategies to include prior knowledge in omics analysis with deep neural networks

High-throughput molecular profiling technologies have revolutionized molecular biology research in the past decades. One important use of molecular data is to make predictions of phenotypes and other features of the organisms using machine learning algorithms. Deep learning models have become increasingly popular for this task due to their ability to learn complex non-linear patterns. Applying deep learning to molecular profiles, however, is challenging due to the very high dimensionality of the data and relatively small sample sizes, causing models to overfit. A solution is to incorporate biological prior knowledge to guide the learning algorithm for processing the functionally related input together. This helps regularize the models and improve their generalizability and interpretability. Here, we describe three major strategies proposed to use prior knowledge in deep learning models to make predictions based on molecular profiles. We review the related deep learning architectures, including the major ideas in relatively new graph neural networks.

Deep-Learning-Based Analysis of Electronic Skin Sensing Data

E-skin is an integrated electronic system that can mimic the perceptual ability of human skin. Traditional analysis methods struggle to handle complex e-skin data, which include time series and multiple patterns, especially when dealing with intricate signals and real-time responses. Recently, deep learning techniques, such as the convolutional neural network, recurrent neural network, and transformer methods, provide effective solutions that can automatically extract data features and recognize patterns, significantly improving the analysis of e-skin data. Deep learning is not only capable of handling multimodal data but can also provide real-time response and personalized predictions in dynamic environments. Nevertheless, problems such as insufficient data annotation and high demand for computational resources still limit the application of e-skin. Optimizing deep learning algorithms, improving computational efficiency, and exploring hardware-algorithm co-designing will be the key to future development. This review aims to present the deep learning techniques applied in e-skin and provide inspiration for subsequent researchers. We first summarize the sources and characteristics of e-skin data and review the deep learning models applicable to e-skin data and their applications in data analysis. Additionally, we discuss the use of deep learning in e-skin, particularly in health monitoring and human-machine interactions, and we explore the current challenges and future development directions.

Artificial intelligence-based biomarkers for treatment decisions in oncology

The development of new therapeutic strategies such as immune checkpoint inhibitors (ICIs) and targeted therapies has increased the complexity of the treatment landscape for solid tumors. At the current rate of annual FDA approvals, the potential treatment options could increase by tenfold over the next 5 years. The cost of personalized medicine technologies limits its accessibility, thus increasing socioeconomic disparities in the treated population. In this review we describe artificial intelligence (AI)-based solutions - including deep learning (DL) methods for routine medical imaging and large language models (LLMs) for electronic health records (EHRs) - to support cancer treatment decisions with cost-effective biomarkers. We address the current limitations of these technologies and propose the next steps towards their adoption in routine clinical practice.

Prediction of checkpoint inhibitor immunotherapy efficacy for cancer using routine blood tests and clinical data

Predicting whether a patient with cancer will benefit from immune checkpoint inhibitors (ICIs) without resorting to advanced genomic or immunologic assays is an important clinical need. To address this, we developed and evaluated SCORPIO, a machine learning system that utilizes routine blood tests (complete blood count and comprehensive metabolic profile) alongside clinical characteristics from 9,745 ICI-treated patients across 21 cancer types. SCORPIO was trained on data from 1,628 patients across 17 cancer types from Memorial Sloan Kettering Cancer Center. In two internal test sets comprising 2,511 patients across 19 cancer types, SCORPIO achieved median time-dependent area under the receiver operating characteristic curve (AUC(t)) values of 0.763 and 0.759 for predicting overall survival at 6, 12, 18, 24 and 30 months, outperforming tumor mutational burden (TMB), which showed median AUC(t) values of 0.503 and 0.543. Additionally, SCORPIO demonstrated superior predictive performance for predicting clinical benefit (tumor response or prolonged stability), with AUC values of 0.714 and 0.641, compared to TMB (AUC = 0.546 and 0.573). External validation was performed using 10 global phase 3 trials (4,447 patients across 6 cancer types) and a real-world cohort from the Mount Sinai Health System (1,159 patients across 18 cancer types). In these external cohorts, SCORPIO maintained robust performance in predicting ICI outcomes, surpassing programmed death-ligand 1 immunostaining. These findings underscore SCORPIO's reliability and adaptability, highlighting its potential to predict patient outcomes with ICI therapy across diverse cancer types and healthcare settings.

Towards an interpretable deep learning model of cancer

Cancer is a manifestation of dysfunctional cell states. It emerges from an interplay of intrinsic and extrinsic factors that disrupt cellular dynamics, including genetic and epigenetic alterations, as well as the tumor microenvironment. This complexity can make it challenging to infer molecular causes for treating the disease. This may be addressed by system-wide computer models of cells, as they allow rapid generation and testing of hypotheses that would be too slow or impossible to perform in the laboratory and clinic. However, so far, such models have been impeded by both experimental and computational limitations. In this perspective, we argue that they can now be achieved using deep learning algorithms to integrate omics data and prior knowledge of molecular networks. Such models would have many applications in precision oncology, e.g., for identifying drug targets and biomarkers, predicting resistance mechanisms and toxicity effects of drugs, or simulating cell-cell interactions in the microenvironment.

A hybrid machine learning framework for functional annotation of mitochondrial glutathione transport and metabolism proteins in cancers

BACKGROUND

Alterations of metabolism, including changes in mitochondrial metabolism as well as glutathione (GSH) metabolism are a well appreciated hallmark of many cancers. Mitochondrial GSH (mGSH) transport is a poorly characterized aspect of GSH metabolism, which we investigate in the context of cancer. Existing functional annotation approaches from machine (ML) or deep learning (DL) models based only on protein sequences, were unable to annotate functions in biological contexts.

RESULTS

We develop a flexible ML framework for functional annotation from diverse feature data. This hybrid ML framework leverages cancer cell line multi-omics data and other biological knowledge data as features, to uncover potential genes involved in mGSH metabolism and membrane transport in cancers. This framework achieves strong performance across functional annotation tasks and several cell line and primary tumor cancer samples. For our application, classification models predict the known mGSH transporter SLC25A39 but not SLC25A40 as being highly probably related to mGSH metabolism in cancers. SLC25A10, SLC25A50, and orphan SLC25A24, SLC25A43 are predicted to be associated with mGSH metabolism in multiple biological contexts and structural analysis of these proteins reveal similarities in potential substrate binding regions to the binding residues of SLC25A39.

CONCLUSION

These findings have implications for a better understanding of cancer cell metabolism and novel therapeutic targets with respect to GSH metabolism through potential novel functional annotations of genes. The hybrid ML framework proposed here can be applied to other biological function classifications or multi-omics datasets to generate hypotheses in various biological contexts. Code and a tutorial for generating models and predictions in this framework are available at: https://github.com/lkenn012/mGSH_cancerClassifiers .

The Hallmarks of Predictive Oncology

SIGNIFICANCE

As the field of artificial intelligence evolves rapidly, these hallmarks are intended to capture fundamental, complementary concepts necessary for the progress and timely adoption of predictive modeling in precision oncology. Through these hallmarks, we hope to establish standards and guidelines that enable the symbiotic development of artificial intelligence and precision oncology.

Optimizing Cancer Treatment: Exploring the Role of AI in Radioimmunotherapy

Radioimmunotherapy (RIT) is a novel cancer treatment that combines radiotherapy and immunotherapy to precisely target tumor antigens using monoclonal antibodies conjugated with radioactive isotopes. This approach offers personalized, systemic, and durable treatment, making it effective in cancers resistant to conventional therapies. Advances in artificial intelligence (AI) present opportunities to enhance RIT by improving precision, efficiency, and personalization. AI plays a critical role in patient selection, treatment planning, dosimetry, and response assessment, while also contributing to drug design and tumor classification. This review explores the integration of AI into RIT, emphasizing its potential to optimize the entire treatment process and advance personalized cancer care.

A mechanism-informed deep neural network enables prioritization of regulators that drive cell state transitions

Cells are regulated at multiple levels, from regulations of individual genes to interactions across multiple genes. Some recent neural network models can connect molecular changes to cellular phenotypes, but their design lacks modeling of regulatory mechanisms, limiting the decoding of regulations behind key cellular events, such as cell state transitions. Here, we present regX, a deep neural network incorporating both gene-level regulation and gene-gene interaction mechanisms, which enables prioritizing potential driver regulators of cell state transitions and providing mechanistic interpretations. Applied to single-cell multi-omics data on type 2 diabetes and hair follicle development, regX reliably prioritizes key transcription factors and candidate cis-regulatory elements that drive cell state transitions. Some regulators reveal potential new therapeutic targets, drug repurposing possibilities, and putative causal single nucleotide polymorphisms. This method to analyze single-cell multi-omics data demonstrates how the interpretable design of neural networks can better decode biological systems.

M-NET: Transforming Single Nucleotide Variations Into Patient Feature Images for the Prediction of Prostate Cancer Metastasis and Identification of Significant Pathways

High-performance prediction of prostate cancer metastasis based on single nucleotide variations remains a challenge. Therefore, we developed a novel biologically informed deep learning framework, named M-NET, for the prediction of prostate cancer metastasis. Within the framework, we transformed single nucleotide variations into patient feature images that are optimal for fitting convolutional neural networks. Moreover, we identified significant pathways associated with the metastatic status. The experimental results showed that M-NET significantly outperformed other comparison methods based on single nucleotide variations, achieving improvements in accuracy, precision, recall, F1-score, area under the receiver operating characteristics curve, and area under the precision-recall curve by 6.3%, 8.4%, 5.1%, 0.070, 0.041, and 0.026, respectively. Furthermore, M-NET identified some important pathways associated with the metastatic status, such as signaling by the hedgehog pathway. In summary, compared with other comparative methods, M-NET exhibited a better performance in the prediction of prostate cancer metastasis.

Personalized Medicine: Leave no Patient Behind; Report From the 2024 Coffey-Holden Prostate Cancer Academy Meeting

INTRODUCTION

The 11th Annual 2024 Coffey - Holden Prostate Cancer Academy (CHPCA) Meeting, was themed "Personalized Medicine: Leave No Patient Behind," and was held from June 20 to 23, 2024 at the University of California, Los Angeles, Luskin Conference Center, in Los Angeles, CA.

METHODS

The CHPCA Meeting is an academy-styled annual conference organized by the Prostate Cancer Foundation, to focus discussion on the most critical emerging research that have the greatest potential to advance knowledge of prostate cancer biology and treatment. The 2024 CHPCA Meeting was attended by 75 academic investigators and included 37 talks across 8 sessions.

RESULTS

The meeting sessions focused on: novel human, mouse and systems biology research models, novel immunotherapies for prostate cancer, efforts to overcome treatment resistance, the role of metabolism and diet in prostate cancer biology and as a therapeutic target, mechanisms that drive differentiation into neuroendocrine cancer subtypes, the evolving prostate cancer epigenome in disease progression and treatment resistance, and machine learning and advanced computational approaches for precision oncology.

DISCUSSION

This article summarizes the presentations and discussions from the 2024 CHPCA Meeting. We hope that sharing this knowledge will inspire and accelerate research into new discoveries and solutions for prostate cancer.

Modeling gene interactions in polygenic prediction via geometric deep learning

Polygenic risk score (PRS) is a widely used approach for predicting individuals' genetic risk of complex diseases, playing a pivotal role in advancing precision medicine. Traditional PRS methods, predominantly following a linear structure, often fall short in capturing the intricate relationships between genotype and phenotype. In this study, we present PRS-Net, an interpretable geometric deep learning-based framework that effectively models the nonlinearity of biological systems for enhanced disease prediction and biological discovery. PRS-Net begins by deconvoluting the genome-wide PRS at the single-gene resolution and then explicitly encapsulates gene-gene interactions leveraging a graph neural network (GNN) for genetic risk prediction, enabling a systematic characterization of molecular interplay underpinning diseases. An attentive readout module is introduced to facilitate model interpretation. Extensive tests across multiple complex traits and diseases demonstrate the superior prediction performance of PRS-Net compared with a wide range of conventional PRS methods. The interpretability of PRS-Net further enhances the identification of disease-relevant genes and gene programs. PRS-Net provides a potent tool for concurrent genetic risk prediction and biological discovery for complex diseases.

Genome-wide association study on color-image-based convolutional neural networks

Background

Convolutional neural networks have excellent modeling abilities to complex large-scale datasets and have been applied to genomics. It requires converting genotype data to image format when employing convolutional neural networks to genome-wide association studies. Existing studies converting the data into grayscale images have shown promising. However, the grayscale image may cause the loss of information of the genotype data.

Methods

In order to make full use of the information, we proposed a new method, color-image-based convolutional neural networks, by converting the data into color images.

Results

The experiments on simulation and real data show that our method outperforms the existing methods proposed by Yue and Chen for converting data into grayscale images, in which the model accuracy is improved by an average of 7.61%, and the ratio of disease risk genes is increased by an average of 18.91%. The new method has better robustness and generalized performance.

G2PDeep-v2: a web-based deep-learning framework for phenotype prediction and biomarker discovery for all organisms using multi-omics data

The G2PDeep-v2 server is a web-based platform powered by deep learning, for phenotype prediction and markers discovery from multi-omics data in any organisms including humans, plants, animals, and viruses. The server provides multiple services for researchers to create deep-learning models through an interactive interface and train these models using an automated hyperparameter tuning algorithm on high-performance computing resources. Users can visualize the results of phenotype and markers predictions and perform Gene Set Enrichment Analysis for the significant markers to provide insights into the molecular mechanisms underlying complex diseases, conditions and other biological phenotypes being studied. The G2PDeep-v2 server is publicly available at https://g2pdeep.org/ and can be utilized for all organisms.

Deep Learning for Biomarker Discovery in Cancer Genomes

Background

Genomic data is essential for clinical decision-making in precision oncology. Bioinformatic algorithms are widely used to analyze next-generation sequencing (NGS) data, but they face two major challenges. First, these pipelines are highly complex, involving multiple steps and the integration of various tools. Second, they generate features that are human-interpretable but often result in information loss by focusing only on predefined genetic properties. This limitation restricts the full potential of NGS data in biomarker extraction and slows the discovery of new biomarkers in precision oncology.

Methods

We propose an end-to-end deep learning (DL) approach for analyzing NGS data. Specifically, we developed a multiple instance learning DL framework that integrates somatic mutation sequences to predict two compound biomarkers: microsatellite instability (MSI) and homologous recombination deficiency (HRD). To achieve this, we utilized data from 3,184 cancer patients obtained from two public databases: The Cancer Genome Atlas (TCGA) and the Clinical Proteome Tumor Analysis Consortium (CPTAC).

Results

Our proposed deep learning method demonstrated high accuracy in identifying clinically relevant biomarkers. For predicting MSI status, the model achieved an accuracy of 0.98, a sensitivity of 0.95, and a specificity of 1.00 on an external validation cohort. For predicting HRD status, the model achieved an accuracy of 0.80, a sensitivity of 0.75, and a specificity of 0.86. Furthermore, the deep learning approach significantly outperformed traditional machine learning methods in both tasks (MSI accuracy, p-value = 5.11×10-18; HRD accuracy, p-value = 1.07×10-10). Using explainability techniques, we demonstrated that the model's predictions are based on biologically meaningful features, aligning with key DNA damage repair mutation signatures.

Conclusion

We demonstrate that deep learning can identify patterns in unfiltered somatic mutations without the need for manual feature extraction. This approach enhances the detection of actionable targets and paves the way for developing NGS-based biomarkers using minimally processed data.

Novel Inhibitors for MDM2-MDM4 E3 Ligase Potently Induce p53-Indepedent Apoptosis in Drug-Resistant Leukemic Cells

MDM2 and MDM4 are major negative regulators of tumor suppressor p53. Beyond regulating p53, MDM2 possesses p53-independent activity in promoting cell cycle progression and tumorigenesis via its RING domain ubiquitin E3 ligase activity. MDM2 and MDM4 form heterodimer polyubiquitin E3 ligases via their RING domain interaction. Inhibitors disrupting p53 interaction with MDM2/MDM4 are in clinical trials in patients bearing wild-type p53 cancers. However, these inhibitors are not designed to work for p53-null/mutant cancer cells. Owing to the importance of the E3 ligase of MDM2 in its p53-independent oncogenic activity, inhibitors targeting the E3 ligase activity of MDM2-MDM4 are desirable for p53-mutant cancer cells. Here, we report the development of such inhibitors with pro-apoptotic activity in p53-null leukemic cells. Among analogues of MDM2-MDM4 E3 ligase inhibitors, we initially identified MMRi36 as a potent pro-apoptotic compound in p53-null leukemic cells with acquired drug resistance. MMRi36 acts as an activator of MDM2-MDM4 E3 ligase by stabilizing MDM2-MDM4 heterodimers and promotes MDM2/MDM4 degradation in cells. Interestingly, replacement of the sulfur in 1,3,4-thiadiazole MMRi36 with a carbon led to identification of pyrazole MMRi36C that dissociates the MDM2-MDM4 RING heterodimers, inhibits the E3 ligase activity of the complex, and induces p53 protein accumulation, but retains the p53-independent pro-apoptotic activity. A brief SAR study identified a fluorine derivative of MMRi36C with improved pro-apoptotic activity. This study discovered a novel class of compound that targets MDM2-MDM4 ubiquitin E3 ligase activity for apoptosis induction in p53-mutant cancer cells.

Advancements in Mass Spectrometry-Based Targeted Metabolomics and Lipidomics: Implications for Clinical Research

Targeted metabolomics and lipidomics are increasingly utilized in clinical research, providing quantitative and comprehensive assessments of metabolic profiles that underlie physiological and pathological mechanisms. These approaches enable the identification of critical metabolites and metabolic alterations essential for accurate diagnosis and precision treatment. Mass spectrometry, in combination with various separation techniques, offers a highly sensitive and specific platform for implementing targeted metabolomics and lipidomics in clinical settings. Nevertheless, challenges persist in areas such as sample collection, quantification, quality control, and data interpretation. This review summarizes recent advances in targeted metabolomics and lipidomics, emphasizing their applications in clinical research. Advancements, including microsampling, dynamic multiple reaction monitoring, and integration of ion mobility mass spectrometry, are highlighted. Additionally, the review discusses the critical importance of data standardization and harmonization for successful clinical implementation.

Artificial intelligence-aided data mining of medical records for cancer detection and screening

The application of artificial intelligence methods to electronic patient records paves the way for large-scale analysis of multimodal data. Such population-wide data describing deep phenotypes composed of thousands of features are now being leveraged to create data-driven algorithms, which in turn has led to improved methods for early cancer detection and screening. Remaining challenges include establishment of infrastructures for prospective testing of such methods, ways to assess biases given the data, and gathering of sufficiently large and diverse datasets that reflect disease heterogeneities across populations. This Review provides an overview of artificial intelligence methods designed to detect cancer early, including key aspects of concern (eg, the problem of data drift-when the underlying health-care data change over time), ethical aspects, and discrepancies between access to cancer screening in high-income countries versus low-income and middle-income countries.

From molecular mechanisms of prostate cancer to translational applications: based on multi-omics fusion analysis and intelligent medicine

Prostate cancer is the most common cancer in men worldwide and has a high mortality rate. The complex and heterogeneous development of prostate cancer has become a core obstacle in the treatment of prostate cancer. Simultaneously, the issues of overtreatment in early-stage diagnosis, oligometastasis and dormant tumor recognition, as well as personalized drug utilization, are also specific concerns that require attention in the clinical management of prostate cancer. Some typical genetic mutations have been proved to be associated with prostate cancer's initiation and progression. However, single-omic studies usually are not able to explain the causal relationship between molecular alterations and clinical phenotypes. Exploration from a systems genetics perspective is also lacking in this field, that is, the impact of gene network, the environmental factors, and even lifestyle behaviors on disease progression. At the meantime, current trend emphasizes the utilization of artificial intelligence (AI) and machine learning techniques to process extensive multidimensional data, including multi-omics. These technologies unveil the potential patterns, correlations, and insights related to diseases, thereby aiding the interpretable clinical decision making and applications, namely intelligent medicine. Therefore, there is a pressing need to integrate multidimensional data for identification of molecular subtypes, prediction of cancer progression and aggressiveness, along with perosonalized treatment performing. In this review, we systematically elaborated the landscape from molecular mechanism discovery of prostate cancer to clinical translational applications. We discussed the molecular profiles and clinical manifestations of prostate cancer heterogeneity, the identification of different states of prostate cancer, as well as corresponding precision medicine practices. Taking multi-omics fusion, systems genetics, and intelligence medicine as the main perspectives, the current research results and knowledge-driven research path of prostate cancer were summarized.

An interpretable survival model for diffuse large B-cell lymphoma patients using a biologically informed visible neural network

Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin lymphoma (NHL) and is characterized by high heterogeneity. Assessment of its prognosis and genetic subtyping hold significant clinical implications. However, existing DLBCL prognostic models are mainly based on transcriptomic profiles, while genetic variation detection is more commonly used in clinical practice. In addition, current clustering-based subtyping methods mostly focus on genes with high mutation frequencies, providing insufficient explanations for the heterogeneity of DLBCL. Here, we proposed VNNSurv (https://bio-web1.nscc-gz.cn/app/VNNSurv), a survival model for DLBCL patients based on a biologically informed visible neural network (VNN). VNNSurv achieved an average C-index of 0.72 on the cross-validation set (HMRN cohort, n = 928), outperforming the baseline methods. The remarkable interpretability of VNNSurv facilitated the identification of the most impactful genes and the underlying pathways through which they act on patient outcomes. When only the 30 highest-impact genes were used as genetic input, the overall performance of VNNSurv improved, and a C-index of 0.70 was achieved on the external TCGA cohort (n = 48). Leveraging these high-impact genes, including 16 genes with low (<5 %) alteration frequencies, we devised a genetic-based prognostic index (GPI) for risk stratification and a subtype identification method. We stratified the patient group according to the International Prognostic Index (IPI) into three risk grades with significant prognostic differences. Furthermore, the defined subtypes exhibited greater prognostic consistency than clustering-based methods. Broadly, VNNSurv is a valuable DLBCL survival model. Its high interpretability has significant value for precision medicine, and its framework is scalable to other diseases.

Deep Learning of radiology-genomics integration for computational oncology: A mini review

In the field of computational oncology, patient status is often assessed using radiology-genomics, which includes two key technologies and data, such as radiology and genomics. Recent advances in deep learning have facilitated the integration of radiology-genomics data, and even new omics data, significantly improving the robustness and accuracy of clinical predictions. These factors are driving artificial intelligence (AI) closer to practical clinical applications. In particular, deep learning models are crucial in identifying new radiology-genomics biomarkers and therapeutic targets, supported by explainable AI (xAI) methods. This review focuses on recent developments in deep learning for radiology-genomics integration, highlights current challenges, and outlines some research directions for multimodal integration and biomarker discovery of radiology-genomics or radiology-omics that are urgently needed in computational oncology.

DeePathNet: A Transformer-Based Deep Learning Model Integrating Multiomic Data with Cancer Pathways

SIGNIFICANCE

DeePathNet integrates cancer-specific biological pathways using transformer-based deep learning for enhanced cancer analysis. It outperforms existing models in predicting drug responses, cancer types, and subtypes. By enabling pathway-level biomarker discovery, DeePathNet represents a significant advancement in cancer research and could lead to more effective treatments.

π-HuB: the proteomic navigator of the human body

The human body contains trillions of cells, classified into specific cell types, with diverse morphologies and functions. In addition, cells of the same type can assume different states within an individual's body during their lifetime. Understanding the complexities of the proteome in the context of a human organism and its many potential states is a necessary requirement to understanding human biology, but these complexities can neither be predicted from the genome, nor have they been systematically measurable with available technologies. Recent advances in proteomic technology and computational sciences now provide opportunities to investigate the intricate biology of the human body at unprecedented resolution and scale. Here we introduce a big-science endeavour called π-HuB (proteomic navigator of the human body). The aim of the π-HuB project is to (1) generate and harness multimodality proteomic datasets to enhance our understanding of human biology; (2) facilitate disease risk assessment and diagnosis; (3) uncover new drug targets; (4) optimize appropriate therapeutic strategies; and (5) enable intelligent healthcare, thereby ushering in a new era of proteomics-driven phronesis medicine. This ambitious mission will be implemented by an international collaborative force of multidisciplinary research teams worldwide across academic, industrial and government sectors.

Immune digital twins for complex human pathologies: applications, limitations, and challenges

Digital twins represent a key technology for precision health. Medical digital twins consist of computational models that represent the health state of individual patients over time, enabling optimal therapeutics and forecasting patient prognosis. Many health conditions involve the immune system, so it is crucial to include its key features when designing medical digital twins. The immune response is complex and varies across diseases and patients, and its modelling requires the collective expertise of the clinical, immunology, and computational modelling communities. This review outlines the initial progress on immune digital twins and the various initiatives to facilitate communication between interdisciplinary communities. We also outline the crucial aspects of an immune digital twin design and the prerequisites for its implementation in the clinic. We propose some initial use cases that could serve as "proof of concept" regarding the utility of immune digital technology, focusing on diseases with a very different immune response across spatial and temporal scales (minutes, days, months, years). Lastly, we discuss the use of digital twins in drug discovery and point out emerging challenges that the scientific community needs to collectively overcome to make immune digital twins a reality.

XModNN: Explainable Modular Neural Network to Identify Clinical Parameters and Disease Biomarkers in Transcriptomic Datasets

The Explainable Modular Neural Network (XModNN) enables the identification of biomarkers, facilitating the classification of diseases and clinical parameters in transcriptomic datasets. The modules within XModNN represent specific pathways or genes of a functional hierarchy. The incorporation of biological insights into the architectural design reduced the number of parameters. This is further reinforced by the weighted multi-loss progressive training, which enables successful classification with a reduced number of replicates. The combination of this workflow with layer-wise relevance propagation ensures a robust post hoc explanation of the individual module contribution. Two use cases were employed to predict sex and neuroblastoma cell states, demonstrating that XModNN, in contrast to standard statistical approaches, results in a reduced number of candidate biomarkers. Moreover, the architecture enables the training on a limited number of examples, attaining the same performance and robustness as support vector machine and random forests. The integrated pathway relevance analysis improves a standard gene set overrepresentation analysis, which relies solely on gene assignment. Two crucial genes and three pathways were identified for sex classification, while 26 genes and six pathways are highly important to discriminate adrenergic-mesenchymal cell states in neuroblastoma cancer.

iGTP: Learning interpretable cellular embedding for inferring biological mechanisms underlying single-cell transcriptomics

Deep-learning models like Variational AutoEncoder have enabled low dimensional cellular embedding representation for large-scale single-cell transcriptomes and shown great flexibility in downstream tasks. However, biologically meaningful latent space is usually missing if no specific structure is designed. Here, we engineered a novel interpretable generative transcriptional program (iGTP) framework that could model the importance of transcriptional program (TP) space and protein-protein interactions (PPI) between different biological states. We demonstrated the performance of iGTP in a diverse biological context using gene ontology, canonical pathway, and different PPI curation. iGTP not only elucidated the ground truth of cellular responses but also surpassed other deep learning models and traditional bioinformatics methods in functional enrichment tasks. By integrating the latent layer with a graph neural network framework, iGTP could effectively infer cellular responses to perturbations. Lastly, we applied iGTP TP embeddings with a latent diffusion model to accurately generate cell embeddings for specific cell types and states. We anticipate that iGTP will offer insights at both PPI and TP levels and holds promise for predicting responses to novel perturbations.

scGO: interpretable deep neural network for cell status annotation and disease diagnosis

Machine learning has emerged as a transformative tool for elucidating cellular heterogeneity in single-cell RNA sequencing. However, a significant challenge lies in the "black box" nature of deep learning models, which obscures the decision-making process and limits interpretability in cell status annotation. In this study, we introduced scGO, a Gene Ontology (GO)-inspired deep learning framework designed to provide interpretable cell status annotation for scRNA-seq data. scGO employs sparse neural networks to leverage the intrinsic biological relationships among genes, transcription factors, and GO terms, significantly augmenting interpretability and reducing computational cost. scGO outperforms state-of-the-art methods in the precise characterization of cell subtypes across diverse datasets. Our extensive experimentation across a spectrum of scRNA-seq datasets underscored the remarkable efficacy of scGO in disease diagnosis, prediction of developmental stages, and evaluation of disease severity and cellular senescence status. Furthermore, we incorporated in silico individual gene manipulations into the scGO model, introducing an additional layer for discovering therapeutic targets. Our results provide an interpretable model for accurately annotating cell status, capturing latent biological knowledge, and informing clinical practice.

c-Triadem: A constrained, explainable deep learning model to identify novel biomarkers in Alzheimer's disease

Alzheimer's disease (AD) is a neurodegenerative disorder that requires early diagnosis for effective management. However, issues with currently available diagnostic biomarkers preclude early diagnosis, necessitating the development of alternative biomarkers and methods, such as blood-based diagnostics. We propose c-Triadem (constrained triple-input Alzheimer's disease model), a novel deep neural network to identify potential blood-based biomarkers for AD and predict mild cognitive impairment (MCI) and AD with high accuracy. The model utilizes genotyping data, gene expression data, and clinical information to predict the disease status of participants, i.e., cognitively normal (CN), MCI, or AD. The nodes of the neural network represent genes and their related pathways, and the edges represent known relationships among the genes and pathways. We trained the model with blood genotyping data, microarray, and clinical features from the Alzheimer's Neuroimaging Disease Initiative (ADNI). We demonstrate that our model's performance is superior to previous models with an AUC of 97% and accuracy of 89%. We then identified the most influential genes and clinical features for prediction using SHapley Additive exPlanations (SHAP). Our SHAP analysis shows that CASP9, LCK, and SDC3 SNPs and PINK1, ATG5, and ubiquitin (UBB, UBC) expression have a higher impact on model performance. Our model has facilitated the identification of potential blood-based genetic markers of DNA damage response and mitophagy in affected regions of the brain. The model can be used for detection and biomarker identification in other related dementias.

Recent Advances in Artificial Sensory Neurons: Biological Fundamentals, Devices, Applications, and Challenges

Spike-based neural networks, which use spikes or action potentials to represent information, have gained a lot of attention because of their high energy efficiency and low power consumption. To fully leverage its advantages, converting the external analog signals to spikes is an essential prerequisite. Conventional approaches including analog-to-digital converters or ring oscillators, and sensors suffer from high power and area costs. Recent efforts are devoted to constructing artificial sensory neurons based on emerging devices inspired by the biological sensory system. They can simultaneously perform sensing and spike conversion, overcoming the deficiencies of traditional sensory systems. This review summarizes and benchmarks the recent progress of artificial sensory neurons. It starts with the presentation of various mechanisms of biological signal transduction, followed by the systematic introduction of the emerging devices employed for artificial sensory neurons. Furthermore, the implementations with different perceptual capabilities are briefly outlined and the key metrics and potential applications are also provided. Finally, we highlight the challenges and perspectives for the future development of artificial sensory neurons.

An interpretable artificial intelligence framework for designing synthetic lethality-based anti-cancer combination therapies

INTRODUCTION

Synthetic lethality (SL) provides an opportunity to leverage different genetic interactions when designing synergistic combination therapies. To further explore SL-based combination therapies for cancer treatment, it is important to identify and mechanistically characterize more SL interactions. Artificial intelligence (AI) methods have recently been proposed for SL prediction, but the results of these models are often not interpretable such that deriving the underlying mechanism can be challenging.

OBJECTIVES

This study aims to develop an interpretable AI framework for SL prediction and subsequently utilize it to design SL-based synergistic combination therapies.

METHODS

We propose a knowledge and data dual-driven AI framework for SL prediction (KDDSL). Specifically, we use gene knowledge related to the SL mechanism to guide the construction of the model and develop a method to identify the most relevant gene knowledge for the predicted results.

RESULTS

Experimental and literature-based validation confirmed a good balance between predictive and interpretable ability when using KDDSL. Moreover, we demonstrated that KDDSL could help to discover promising drug combinations and clarify associated biological processes, such as the combination of MDM2 and CDK9 inhibitors, which exhibited significant anti-cancer effects in vitro and in vivo.

CONCLUSION

These data underscore the potential of KDDSL to guide SL-based combination therapy design. There is a need for biomedicine-focused AI strategies to combine rational biological knowledge with developed models.

A deep learning model for prediction of autism status using whole-exome sequencing data

Autism is a developmental disability. Research demonstrated that children with autism benefit from early diagnosis and early intervention. Genetic factors are considered major contributors to the development of autism. Machine learning (ML), including deep learning (DL), has been evaluated in phenotype prediction, but this method has been limited in its application to autism. We developed a DL model, the Separate Translated Autism Research Neural Network (STAR-NN) model to predict autism status. The model was trained and tested using whole exome sequencing data from 43,203 individuals (16,809 individuals with autism and 26,394 non-autistic controls). Polygenic scores from common variants and the aggregated count of rare variants on genes were used as input. In STAR-NN, protein truncating variants, possibly damaging missense variants and mild effect missense variants on the same gene were separated at the input level and merged to one gene node. In this way, rare variants with different level of pathogenic effects were treated separately. We further validated the performance of STAR-NN using an independent dataset, including 13,827 individuals with autism and 14,052 non-autistic controls. STAR-NN achieved a modest ROC-AUC of 0.7319 on the testing dataset and 0.7302 on the independent dataset. STAR-NN outperformed other traditional ML models. Gene Ontology analysis on the selected gene features showed an enrichment for potentially informative pathways including calcium ion transport.

Ten challenges and opportunities in computational immuno-oncology

Immuno-oncology has transformed the treatment of cancer, with several immunotherapies becoming the standard treatment across histologies. Despite these advancements, the majority of patients do not experience durable clinical benefits, highlighting the imperative for ongoing advancement in immuno-oncology. Computational immuno-oncology emerges as a forefront discipline that draws on biomedical data science and intersects with oncology, immunology, and clinical research, with the overarching goal to accelerate the development of effective and safe immuno-oncology treatments from the laboratory to the clinic. In this review, we outline 10 critical challenges and opportunities in computational immuno-oncology, emphasizing the importance of robust computational strategies and interdisciplinary collaborations amid the constantly evolving interplay between clinical needs and technological innovation.

Biologically informed deep neural networks provide quantitative assessment of intratumoral heterogeneity in post treatment glioblastoma

Intratumoral heterogeneity poses a significant challenge to the diagnosis and treatment of recurrent glioblastoma. This study addresses the need for non-invasive approaches to map heterogeneous landscape of histopathological alterations throughout the entire lesion for each patient. We developed BioNet, a biologically-informed neural network, to predict regional distributions of two primary tissue-specific gene modules: proliferating tumor (Pro) and reactive/inflammatory cells (Inf). BioNet significantly outperforms existing methods (p < 2e-26). In cross-validation, BioNet achieved AUCs of 0.80 (Pro) and 0.81 (Inf), with accuracies of 80% and 75%, respectively. In blind tests, BioNet achieved AUCs of 0.80 (Pro) and 0.76 (Inf), with accuracies of 81% and 74%. Competing methods had AUCs lower or around 0.6 and accuracies lower or around 70%. BioNet's voxel-level prediction maps reveal intratumoral heterogeneity, potentially improving biopsy targeting and treatment evaluation. This non-invasive approach facilitates regular monitoring and timely therapeutic adjustments, highlighting the role of ML in precision medicine.

E-Skin and Its Advanced Applications in Ubiquitous Health Monitoring

E-skin is a bionic device with flexible and intelligent sensing ability that can mimic the touch, temperature, pressure, and other sensing functions of human skin. Because of its flexibility, breathability, biocompatibility, and other characteristics, it is widely used in health management, personalized medicine, disease prevention, and other pan-health fields. With the proposal of new sensing principles, the development of advanced functional materials, the development of microfabrication technology, and the integration of artificial intelligence and algorithms, e-skin has developed rapidly. This paper focuses on the characteristics, fundamentals, new principles, key technologies, and their specific applications in health management, exercise monitoring, emotion and heart monitoring, etc. that advanced e-skin needs to have in the healthcare field. In addition, its significance in infant and child care, elderly care, and assistive devices for the disabled is analyzed. Finally, the current challenges and future directions of the field are discussed. It is expected that this review will generate great interest and inspiration for the development and improvement of novel e-skins and advanced health monitoring systems.

Transparent sparse graph pathway network for analyzing the internal relationship of lung cancer

While it is important to find the key biomarkers and improve the accuracy of disease models, it is equally important to understand their interaction relationships. In this study, a transparent sparse graph pathway network (TSGPN) is proposed based on the structure of graph neural networks. This network simulates the action of genes in vivo, adds to prior knowledge, and improves the model's accuracy. First, the graph connection was constructed according to protein-protein interaction networks and competing endogenous RNA (ceRNA) networks, from which some noise or unimportant connections were spontaneously removed based on the graph attention mechanism and hard concrete estimation. This realized the reconstruction of the ceRNA network representing the influence of other genes in the disease on mRNA. Next, the gene-based interpretation was transformed into a pathway-based interpretation based on the pathway database, and the hidden layer was added to realize the high-dimensional analysis of the pathway. Finally, the experimental results showed that the proposed TSGPN method is superior to other comparison methods in F1 score and AUC, and more importantly, it can effectively display the role of genes. Through data analysis applied to lung cancer prognosis, ten pathways related to LUSC prognosis were found, as well as the key biomarkers closely related to these pathways, such as HOXA10, hsa-mir-182, and LINC02544. The relationship between them was also reconstructed to better explain the internal mechanism of the disease.

G2PDeep-v2: a web-based deep-learning framework for phenotype prediction and biomarker discovery using multi-omics data

The G2PDeep-v2 server is a web-based platform powered by deep learning, for phenotype prediction and markers discovery from multi-omics data in any organisms including humans, plants, animals, and viruses. The server provides multiple services for researchers to create deep-learning models through an interactive interface and train these models using an automated hyperparameter tuning algorithm on high-performance computing resources. Users can visualize the results of phenotype and markers predictions and perform Gene Set Enrichment Analysis for the significant markers to provide insights into the molecular mechanisms underlying complex diseases and other biological processes. The G2PDeep-v2 server is publicly available at https://g2pdeep.org/.

Cellular, Structural Basis, and Recent Progress for Targeting Murine Double Minute X (MDMX) in Tumors

Murine double minute X (MDMX) is an oncoprotein that mainly has a negative regulatory effect on the tumor suppressor p53 to induce tumorigenesis. As MDMX is highly expressed in various types of tumor cells, targeting and inhibiting MDMX are becoming a promising strategy for treating cancers. However, the high degree of structural homology between MDMX and its homologous protein murine double minute 2 (MDM2) is a great challenge for the development of MDMX-targeted therapies. This review introduces the structure, distribution, and regulation of the MDMX, summarizes the structural features and structure-activity relationships (SARs) of MDMX ligands, and focuses on the differences between MDMX and MDM2 in these aspects. Our purpose of this work is to propose potential strategies to achieve the specific targeting of MDMX.

Molecule Design for Non-Aqueous Wide-Temperature Electrolytes via the Intelligentized Screening Method

Operating a lithium-ion battery (LIB) in a wide temperature range is essential for ensuring a stable electricity supply amidst fluctuating temperatures caused by climate or terrain changes. Electrolyte plays a pivotal role in determining the temperature durability of batteries. However, specialized electrolytes designed for either low or high temperatures typically possess distinct features. Therefore, wide-temperature electrolytes (WTEs) are necessary as they encompass a combination of diverse properties, which complicates the clear instruction of WTE design. Here we represent an artificial intelligence (Al)-assisted workflow of WTE design through stepwise parameterizations and calculations. Linear mono-nitriles are identified as ideal wide-liquidus-range solvents that can "softly" solvate lithium ions by weak interactions. In addition, the explainable modules revealed the halogenoid similarity of cyanide as fluorine on the electrolyte properties (e.g. boiling point and dielectric constant). With the further introduction of an ether bond, 3-methoxypropionitrile (MPN) has been eventually determined as a main electrolyte solvent, enabling the battery operation from -60 to 120 °C. Particularly, a LiCoO2/Li cell using the proposed WTE can realize stable cycling with capacity retention reaching 72.3 % after 50 cycles under a high temperature of 100 °C.