RESUMEN
Clinically and biologically valuable information may reside untapped in large cancer gene expression data sets. Deep unsupervised learning has the potential to extract this information with unprecedented efficacy but has thus far been hampered by a lack of biological interpretability and robustness. Here, we present DeepProfile, a comprehensive framework that addresses current challenges in applying unsupervised deep learning to gene expression profiles. We use DeepProfile to learn low-dimensional latent spaces for 18 human cancers from 50,211 transcriptomes. DeepProfile outperforms existing dimensionality reduction methods with respect to biological interpretability. Using DeepProfile interpretability methods, we show that genes that are universally important in defining the latent spaces across all cancer types control immune cell activation, while cancer type-specific genes and pathways define molecular disease subtypes. By linking DeepProfile latent variables to secondary tumor characteristics, we discover that tumor mutation burden is closely associated with the expression of cell cycle-related genes. DNA mismatch repair and MHC class II antigen presentation pathway expression, on the other hand, are consistently associated with patient survival. We validate these results through Kaplan-Meier analyses and nominate tumor-associated macrophages as an important source of survival-correlated MHC class II transcripts. Our results illustrate the power of unsupervised deep learning for discovery of novel cancer biology from existing gene expression data.
RESUMEN
Brown rot disease, caused by Monilinia spp., poses a significant threat to pome and stone fruit crops globally, resulting in substantial economic losses during pre- and post-harvest stages. Monilinia fructigena, M. laxa, and M. fructicola are identified as the key agents responsible for brown rot disease. In this study, we employed the amplified fragment length polymorphism (AFLP) method to assess the genetic diversity of 86 strains of Monilinia spp. isolated from major stone fruit cultivation regions in South Korea. Specifically, strains were collected from Chungcheong, Gangwon, Gyeonggi, Gyeongsang, and Jeolla provinces (-do). A comparative analysis of strain characteristics, such as isolation locations, host plants, and responses to chemical fungicides, was conducted. AFLP phylogenetic classification using 20 primer pairs revealed the presence of three distinct groups, with strains from Jeolla province consistently forming a separate group at a high frequency. Furthermore, M. fructicola was divided into three groups by the AFLP pattern. Principal coordinate analysis and PERMANOVA were applied to compare strain information, such as origin, host, and fungicide sensitivity, revealing significant partition patterns for AFLP according to geographic origin and host plants. This study represents the utilization of AFLP methodology to investigate the genetic variability among M. fructicol isolates, highlighting the importance of continuous monitoring and management of variations in the brown rot pathogen.
RESUMEN
Building trustworthy and transparent image-based medical artificial intelligence (AI) systems requires the ability to interrogate data and models at all stages of the development pipeline, from training models to post-deployment monitoring. Ideally, the data and associated AI systems could be described using terms already familiar to physicians, but this requires medical datasets densely annotated with semantically meaningful concepts. In the present study, we present a foundation model approach, named MONET (medical concept retriever), which learns how to connect medical images with text and densely scores images on concept presence to enable important tasks in medical AI development and deployment such as data auditing, model auditing and model interpretation. Dermatology provides a demanding use case for the versatility of MONET, due to the heterogeneity in diseases, skin tones and imaging modalities. We trained MONET based on 105,550 dermatological images paired with natural language descriptions from a large collection of medical literature. MONET can accurately annotate concepts across dermatology images as verified by board-certified dermatologists, competitively with supervised models built on previously concept-annotated dermatology datasets of clinical images. We demonstrate how MONET enables AI transparency across the entire AI system development pipeline, from building inherently interpretable models to dataset and model auditing, including a case study dissecting the results of an AI clinical trial.
Asunto(s)
Inteligencia Artificial , Médicos , Humanos , AprendizajeRESUMEN
Human CD300c is expressed on various immune or cancer cells and is a novel B7 family member, functioning as an activity modulator on immune cells. To elucidate the function of CD300c, we developed CL7, a human CD300c-specific monoclonal antibody, and assessed its biological activity. The specific binding of CL7 monoclonal antibody against recombinant CD300c antigen was confirmed using enzyme-linked immunosorbent assay and surface plasmon resonance analysis. The binding affinity of CL7 was strong at the sub-nanomolar level. Furthermore, CL7 effectively bound to exogenously expressed CD300c on 293T cells. CL7 antibody differentiated monocytes to M1 macrophages, as evidenced by the upregulated expression of M1-specific cell surface markers and increased secretion of M1-specific cytokines in vitro in THP-1 cells and primary macrophages, as well as the increased population size of M1 macrophages in tumors grafted into mice. Additionally, CL7 treatment upregulated PD-L1 expression on THP-1 cells. We confirmed that the mechanism of M1 macrophage differentiation was through the mitogen-activated protein kinase and NF-κB signaling pathways. CD300c expression on various immune and cancer cells was similar to that of the well-known immune checkpoint PD-L1, suggesting the possibility of CD300c as a novel tumor biomarker. We also confirmed that the tumor size was substantially reduced by CL7 antibody treatment in the CT26 mouse model. Our study supports that CD300c is a potential therapeutic target in immuno-oncology. Overall, the CD300c-specific monoclonal antibody, CL7, is a promising immunotherapeutic agent, and it induces enhanced differentiation of M1 macrophages and/or their infiltration into the tumor microenvironment.
Asunto(s)
Antígeno B7-H1 , Monocitos , Humanos , Ratones , Animales , Antígeno B7-H1/metabolismo , Anticuerpos Monoclonales/farmacología , Anticuerpos Monoclonales/metabolismo , Macrófagos , Citocinas/metabolismoRESUMEN
The aim of this study was to investigate the regulation of lantipeptide production in Streptomyces globisporus SP6C4, which produces the novel antifungal lantipeptides conprimycin and grisin, and to identify the role of cytochrome P450 (P450) in tis regulation. To investigate the regulation of lantipeptide production, we created gene deletion mutants, including ΔP450, ΔtsrD, ΔlanM, ΔP450ΔtsrD, and ΔP450ΔlanM. These mutants were characterized in terms of their morphology, sporulation, attachment, and antifungal activity against Fusarium oxysporum. The gene deletion mutants showed distinct characteristics compared to the wild-type strain. Among them, the ΔP450ΔlanM double mutant exhibited a recovery of antifungal activity against F. oxysporum, indicating that P450 plays a significant role in regulating lantipeptide production in S. globisporus SP6C4. Our findings highlight the significant role of P450 in the regulation of lantipeptide production and morphological processes in S. globisporus. The results suggest a potential link between P450-mediated metabolic pathways and the regulation of growth and secondary metabolism in SP6C4, thereby highlighting P450 as a putative target for the development of new antifungal agents.
RESUMEN
The inferences of most machine-learning models powering medical artificial intelligence are difficult to interpret. Here we report a general framework for model auditing that combines insights from medical experts with a highly expressive form of explainable artificial intelligence. Specifically, we leveraged the expertise of dermatologists for the clinical task of differentiating melanomas from melanoma 'lookalikes' on the basis of dermoscopic and clinical images of the skin, and the power of generative models to render 'counterfactual' images to understand the 'reasoning' processes of five medical-image classifiers. By altering image attributes to produce analogous images that elicit a different prediction by the classifiers, and by asking physicians to identify medically meaningful features in the images, the counterfactual images revealed that the classifiers rely both on features used by human dermatologists, such as lesional pigmentation patterns, and on undesirable features, such as background skin texture and colour balance. The framework can be applied to any specialized medical domain to make the powerful inference processes of machine-learning models medically understandable.
RESUMEN
BACKGROUND: Biological age is a measure of health that offers insights into ageing. The existing age clocks, although valuable, often trade off accuracy and interpretability. We introduce ExplaiNAble BioLogical Age (ENABL Age), a computational framework that combines machine-learning models with explainable artificial intelligence (XAI) methods to accurately estimate biological age with individualised explanations. METHODS: To construct the ENABL Age clock, we first predicted an age-related outcome (eg, all-cause or cause-specific mortality), and then rescaled these predictions to estimate biological age, using UK Biobank and National Health and Nutrition Examination Survey (NHANES) datasets. We adapted existing XAI methods to decompose individual ENABL Ages into contributing risk factors. For broad accessibility, we developed two versions: ENABL Age-L, based on blood tests, and ENABL Age-Q, based on questionnaire characteristics. Finally, we validated diverse ageing mechanisms captured by each ENABL Age clock through genome-wide association studies (GWAS) association analyses. FINDINGS: Our ENABL Age clock was significantly correlated with chronological age (r=0·7867, p<0·0001 for UK Biobank; r=0·7126, p<0·0001 for NHANES). These clocks distinguish individuals who are healthy (ie, their ENABL Age is lower than their chronological age) from those who are unhealthy (ie, their ENABL Age is higher than their chronological age), predicting mortality more effectively than existing clocks. Groups of individuals who were unhealthy showed approximately three to 12 times higher log hazard ratio than healthy groups, as per ENABL Age. The clocks achieved high mortality prediction power with an area under the receiver operating characteristic curve of 0·8179 for 5-year mortality and 0·8115 for 10-year mortality on the UK Biobank dataset, and 0·8935 for 5-year mortality and 0·9107 for 10-year mortality on the NHANES dataset. The individualised explanations that revealed the contribution of specific characteristics to ENABL Age provided insights into the important characteristics for ageing. An association analysis with risk factors and ageing-related morbidities and GWAS results on ENABL Age clocks trained on different mortality causes showed that each clock captures distinct ageing mechanisms. INTERPRETATION: ENABL Age brings an important leap forward in the application of XAI for interpreting biological age clocks. ENABL Age also carries substantial potential in practical settings, assisting medical professionals in untangling the complexity of ageing mechanisms, and potentially becoming a valuable tool in informed clinical decision-making processes. FUNDING: National Science Foundation and National Institutes of Health.
Asunto(s)
Inteligencia Artificial , Estudio de Asociación del Genoma Completo , Estados Unidos , Humanos , Encuestas Nutricionales , Aprendizaje Automático , Envejecimiento/genéticaRESUMEN
Single-cell datasets are routinely collected to investigate changes in cellular state between control cells and the corresponding cells in a treatment condition, such as exposure to a drug or infection by a pathogen. To better understand heterogeneity in treatment response, it is desirable to deconvolve variations enriched in treated cells from those shared with controls. However, standard computational models of single-cell data are not designed to explicitly separate these variations. Here, we introduce contrastive variational inference (contrastiveVI; https://github.com/suinleelab/contrastiveVI ), a framework for deconvolving variations in treatment-control single-cell RNA sequencing (scRNA-seq) datasets into shared and treatment-specific latent variables. Using three treatment-control scRNA-seq datasets, we apply contrastiveVI to perform a variety of analysis tasks, including visualization, clustering and differential expression testing. We find that contrastiveVI consistently achieves results that agree with known ground truths and often highlights subtle phenomena that may be difficult to ascertain with standard workflows. We conclude by generalizing contrastiveVI to accommodate joint transcriptome and surface protein measurements.
Asunto(s)
Perfilación de la Expresión Génica , Análisis de la Célula Individual , Perfilación de la Expresión Génica/métodos , Análisis de Secuencia de ARN/métodos , Análisis de la Célula Individual/métodos , Transcriptoma , Análisis por Conglomerados , Algoritmos , Programas InformáticosRESUMEN
Fire blight disease, caused by the bacterial pathogen Erwinia amylovora, has been a significant concern for over 50 countries worldwide. The efficacy of chemical pesticides currently available for disease control is limited. To address this issue, research is being conducted to explore environmentally friendly control methods, particularly biological control using beneficial microorganisms. However, there is limited research on the apple microbiota community and minimal research has been conducted on fungal communities that may exhibit reliable performance in apple trees. Therefore, our objective was to analyze the fungal communities present in apples at different developmental stages and in different tissues, aiming to identify potential biological control agents for fire blight disease. Our findings indicate that the fungal communities present in apple buds, flowers and leaves play an important role in inhibiting the invasion of E. amylovora. Specifically, we propose GS11 and Lipomyces starkeyi as potential keystone taxa that respond to fire blight disease. These findings provide insights into the continuity and discontinuity of fungal community structure in different developmental stages of apples and offer predictions for potential biological control agents for fire blight disease.
RESUMEN
Building trustworthy and transparent image-based medical AI systems requires the ability to interrogate data and models at all stages of the development pipeline: from training models to post-deployment monitoring. Ideally, the data and associated AI systems could be described using terms already familiar to physicians, but this requires medical datasets densely annotated with semantically meaningful concepts. Here, we present a foundation model approach, named MONET (Medical cONcept rETriever), which learns how to connect medical images with text and generates dense concept annotations to enable tasks in AI transparency from model auditing to model interpretation. Dermatology provides a demanding use case for the versatility of MONET, due to the heterogeneity in diseases, skin tones, and imaging modalities. We trained MONET on the basis of 105,550 dermatological images paired with natural language descriptions from a large collection of medical literature. MONET can accurately annotate concepts across dermatology images as verified by board-certified dermatologists, outperforming supervised models built on previously concept-annotated dermatology datasets. We demonstrate how MONET enables AI transparency across the entire AI development pipeline from dataset auditing to model auditing to building inherently interpretable models.
RESUMEN
Despite the proliferation and clinical deployment of artificial intelligence (AI)-based medical software devices, most remain black boxes that are uninterpretable to key stakeholders including patients, physicians, and even the developers of the devices. Here, we present a general model auditing framework that combines insights from medical experts with a highly expressive form of explainable AI that leverages generative models, to understand the reasoning processes of AI devices. We then apply this framework to generate the first thorough, medically interpretable picture of the reasoning processes of machine-learning-based medical image AI. In our synergistic framework, a generative model first renders "counterfactual" medical images, which in essence visually represent the reasoning process of a medical AI device, and then physicians translate these counterfactual images to medically meaningful features. As our use case, we audit five high-profile AI devices in dermatology, an area of particular interest since dermatology AI devices are beginning to achieve deployment globally. We reveal how dermatology AI devices rely both on features used by human dermatologists, such as lesional pigmentation patterns, as well as multiple, previously unreported, potentially undesirable features, such as background skin texture and image color balance. Our study also sets a precedent for the rigorous application of explainable AI to understand AI in any specialized domain and provides a means for practitioners, clinicians, and regulators to uncloak AI's powerful but previously enigmatic reasoning processes in a medically understandable way.
RESUMEN
Epithelial-to-mesenchymal transition (EMT) plays a critical role in the development and progression of lung cancer by promoting its invasiveness and metastasis. Using integrative analyses of the public lung cancer database, we found that the expression levels of the tight junction proteins, zonula occluden (ZO)-1 and ZO-2, were lower in lung cancer tissues, including both lung adenocarcinoma and lung squamous cell carcinoma than in normal lung tissues analyzed using The Cancer Genome Atlas (TCGA). Although the ectopic expression or knockdown of ZO-1 and ZO-2 did not affect the growth of lung cancer cells, they significantly regulated cell migration and invasion. When M0 macrophages were co-cultured with ZO-1 or ZO-2 knockdown Calu-1 cells, M2-like polarization was efficiently induced. Conversely, co-culture of M0 THP-1 cells with A549 cells stably expressing ZO-1 or ZO-2 significantly reduced M2 differentiation. We also identified G protein subunit alpha q (GNAQ) as a potential ZO-1- and ZO-2-specific activator through analysis of correlated genes with the TCGA lung cancer database. Our results suggest that the GNAQ-ZO-1/2 axis may play a tumor-suppressive role in lung cancer development and progression and highlight ZO-1 and ZO-2 as key EMT- and tumor microenvironment-suppressive proteins. These findings provide new insights for the development of targeted therapies for lung cancer.
Asunto(s)
Adenocarcinoma del Pulmón , Neoplasias Pulmonares , Humanos , Uniones Estrechas/metabolismo , Microambiente Tumoral/genética , Neoplasias Pulmonares/genética , Transición Epitelial-Mesenquimal/genética , Proteína de la Zonula Occludens-1/genética , Proteína de la Zonula Occludens-1/metabolismo , Subunidades alfa de la Proteína de Unión al GTP Gq-G11/metabolismoRESUMEN
Machine learning may aid the choice of optimal combinations of anticancer drugs by explaining the molecular basis of their synergy. By combining accurate models with interpretable insights, explainable machine learning promises to accelerate data-driven cancer pharmacology. However, owing to the highly correlated and high-dimensional nature of transcriptomic data, naively applying current explainable machine-learning strategies to large transcriptomic datasets leads to suboptimal outcomes. Here by using feature attribution methods, we show that the quality of the explanations can be increased by leveraging ensembles of explainable machine-learning models. We applied the approach to a dataset of 133 combinations of 46 anticancer drugs tested in ex vivo tumour samples from 285 patients with acute myeloid leukaemia and uncovered a haematopoietic-differentiation signature underlying drug combinations with therapeutic synergy. Ensembles of machine-learning models trained to predict drug combination synergies on the basis of gene-expression data may improve the feature attribution quality of complex machine-learning models.
Asunto(s)
Perfilación de la Expresión Génica , Aprendizaje Automático , Humanos , TranscriptomaRESUMEN
As interest in using unsupervised deep learning models to analyze gene expression data has grown, an increasing number of methods have been developed to make these models more interpretable. These methods can be separated into two groups: post hoc analyses of black box models through feature attribution methods and approaches to build inherently interpretable models through biologically-constrained architectures. We argue that these approaches are not mutually exclusive, but can in fact be usefully combined. We propose PAUSE ( https://github.com/suinleelab/PAUSE ), an unsupervised pathway attribution method that identifies major sources of transcriptomic variation when combined with biologically-constrained neural network models.
Asunto(s)
Perfilación de la Expresión Génica , Transcriptoma , Redes Neurales de la ComputaciónRESUMEN
A prominent trend in single-cell transcriptomics is providing spatial context alongside a characterization of each cell's molecular state. This typically requires targeting an a priori selection of genes, often covering less than 1% of the genome, and a key question is how to optimally determine the small gene panel. We address this challenge by introducing a flexible deep learning framework, PERSIST, to identify informative gene targets for spatial transcriptomics studies by leveraging reference scRNA-seq data. Using datasets spanning different brain regions, species, and scRNA-seq technologies, we show that PERSIST reliably identifies panels that provide more accurate prediction of the genome-wide expression profile, thereby capturing more information with fewer genes. PERSIST can be adapted to specific biological goals, and we demonstrate that PERSIST's binarization of gene expression levels enables models trained on scRNA-seq data to generalize with to spatial transcriptomics data, despite the complex shift between these technologies.
Asunto(s)
Análisis de la Célula Individual , Transcriptoma , Transcriptoma/genética , Perfilación de la Expresión Génica , Análisis de Secuencia de ARNRESUMEN
Background: Unlike linear models which are traditionally used to study all-cause mortality, complex machine learning models can capture non-linear interrelations and provide opportunities to identify unexplored risk factors. Explainable artificial intelligence can improve prediction accuracy over linear models and reveal great insights into outcomes like mortality. This paper comprehensively analyzes all-cause mortality by explaining complex machine learning models. Methods: We propose the IMPACT framework that uses XAI technique to explain a state-of-the-art tree ensemble mortality prediction model. We apply IMPACT to understand all-cause mortality for 1-, 3-, 5-, and 10-year follow-up times within the NHANES dataset, which contains 47,261 samples and 151 features. Results: We show that IMPACT models achieve higher accuracy than linear models and neural networks. Using IMPACT, we identify several overlooked risk factors and interaction effects. Furthermore, we identify relationships between laboratory features and mortality that may suggest adjusting established reference intervals. Finally, we develop highly accurate, efficient and interpretable mortality risk scores that can be used by medical professionals and individuals without medical expertise. We ensure generalizability by performing temporal validation of the mortality risk scores and external validation of important findings with the UK Biobank dataset. Conclusions: IMPACT's unique strength is the explainable prediction, which provides insights into the complex, non-linear relationships between mortality and features, while maintaining high accuracy. Our explainable risk scores could help individuals improve self-awareness of their health status and help clinicians identify patients with high risk. IMPACT takes a consequential step towards bringing contemporary developments in XAI to epidemiology.
This study identifies characteristics that will make a person more likely to die sooner than expected based on life expectancy for the population. We developed a computer program and applied it to information obtained about the characteristics and medical history of people from the USA. We identified previously unidentified characteristics that impact how likely it is someone will die sooner than expected, for example the circumference of the arm. We also identified combinations of characteristics that interact to increase the likelihood of death sooner than expected. By adding a person's characteristics to the program, the likelihood of death over the next 5 years can be calculated and characteristics identified that a person could modify to improve their health and reduce their chance of death during this period.
RESUMEN
Soybean is an important source of protein and for a wide range of agricultural, food, and industrial applications. Soybean is being affected by Xanthomonas citri pv. glycines, a causal pathogen of bacterial pustule disease, result in a reduction in yield and quality. Diverse microbial communities of plants are involved in various plant stresses is known. Therefore, we designed to investigate the microbial community differentiation depending on the infection of X. citri pv. glycines. The microbial community's abundance, diversity, and similarity showed a difference between infected and non-infected soybean. Microbiota community analysis, excluding X. citri pv. glycines, revealed that Pseudomonas spp. would increase the population of the infected soybean. Results of DESeq analyses suggested that energy metabolism, secondary metabolite, and TCA cycle metabolism were actively diverse in the non-infected soybeans. Additionally, Streptomyces bacillaris S8, an endophyte microbiota member, was nominated as a key microbe in the healthy soybeans. Genome analysis of S. bacillaris S8 presented that salinomycin may be the critical antibacterial metabolite. Our findings on the composition of soybean microbiota communities and the key strain information will contribute to developing biological control strategies against X. citri pv. glycines.
RESUMEN
Local feature attribution methods are increasingly used to explain complex machine learning models. However, current methods are limited because they are extremely expensive to compute or are not capable of explaining a distributed series of models where each model is owned by a separate institution. The latter is particularly important because it often arises in finance where explanations are mandated. Here, we present Generalized DeepSHAP (G-DeepSHAP), a tractable method to propagate local feature attributions through complex series of models based on a connection to the Shapley value. We evaluate G-DeepSHAP across biological, health, and financial datasets to show that it provides equally salient explanations an order of magnitude faster than existing model-agnostic attribution techniques and demonstrate its use in an important distributed series of models setting.
Asunto(s)
Aprendizaje AutomáticoRESUMEN
Although knowledge of biological pathways is essential for interpreting results from computational biology studies, the growing number of pathway databases complicates efforts to efficiently perform pathway analysis due to high redundancies among pathways from different databases, and inconsistencies in how pathways are created and named. We introduce the PAthway Communities (PAC) framework, which reconciles pathways from different databases and reduces pathway redundancy by revealing informative groups with distinct biological functions. Uniquely applying the Louvain community detection algorithm to a network of 4847 pathways from KEGG, REACTOME and Gene Ontology databases, we identify 35 distinct and automatically annotated communities of pathways and show that they are consistent with expert-curated pathway categories. Further, we demonstrate that our pathway community network can be queried with new gene sets to provide biological context in terms of related pathways and communities. Our approach, combined with an interpretable web tool we provide, will help computational biologists more efficiently contextualize and interpret their biological findings.