Featured
-
-
Article
| Open AccessUncertainty-inspired open set learning for retinal anomaly identification
Failure to recognize samples from unseen classes is a major limitation of AI recognition and classification of retinal anomalies. Here, the authors present the Uncertainty-inspired Open Set learning model that categorises fundus images into pre-trained categories, and provides an uncertainty score that alerts the need for manual inspection when dealing with out-of-distribution images.
- Meng Wang
- , Tian Lin
- & Huazhu Fu
-
Article
| Open AccessNivolumab and ipilimumab in recurrent or refractory cancer of unknown primary: a phase II trial
Standard of care for unfavorable-risk cancer of unknown primary (CUP) comprises platinum-based chemotherapy as first-line treatment, however therapeutic options remain limited. Here the authors report the results of a phase II trial of combined nivolumab (anti-PD1) and ipilimumab (anti-CTLA4) in patients with unfavorable CUP.
- Maria Pouyiourou
- , Bianca N. Kraft
- & Alwin Krämer
-
Article
| Open AccessPlasma biomarkers predict Alzheimer’s disease before clinical onset in Chinese cohorts
Performance of plasma biomarkers of amyloid (A), tau (T) and neurodegeneration (N) for Alzheimer’s disease (AD) in Chinese cohorts is unknown. Here, the authors report that plasma ATN biomarkers can predict AD 8–10 years before symptoms in Chinese cohorts.
- Huimin Cai
- , Yana Pang
- & Longfei Jia
-
Article
| Open AccessIntegrated radiogenomics models predict response to neoadjuvant chemotherapy in high grade serous ovarian cancer
Response to treatment in high grade serous ovarian carcinoma (HGSOC) is highly variable. Here, the authors leverage a radiogenomic model to predict neoadjuvant chemotherapy response in HGSOC, including clinical data, medical imaging, and blood-based biomarkers such as CA-125 and ctDNA features.
- Mireia Crispin-Ortuzar
- , Ramona Woitek
- & James D. Brenton
-
Article
| Open AccessClinical utility of polygenic scores for cardiometabolic disease in Arabs
Arabs account for 5% of the world population and have a high burden of cardiometabolic disease. Here, the authors optimize polygenic scores for 10 cardiometabolic traits in 5399 Arabs, achieving a performance on par with that among European-ancestry individuals.
- Injeong Shim
- , Hiroyuki Kuwahara
- & Akl C. Fahed
-
Article
| Open AccessMetabolic fingerprinting on retinal pigment epithelium thickness for individualized risk stratification of type 2 diabetes mellitus
The retina’s role in type 2 diabetes is not well understood. Here, the authors show that retinal pigment epithelium thickness, with its metabolic fingerprints, might offer improved predictability and clinical utility beyond traditional indicators.
- Shaopeng Yang
- , Zhuoting Zhu
- & Wei Wang
-
Article
| Open AccessA human lung alveolus-on-a-chip model of acute radiation-induced lung injury
Acute exposure to radiation can lead to acute pneumonitis, fibrosis or death. Here the authors develop an alveolus-on chip model to study the molecular characteristics of radiation induced lung injury, better understand radiation induced lung disease and facilitate drug screening.
- Queeny Dasgupta
- , Amanda Jiang
- & Donald E. Ingber
-
Article
| Open AccessRevealing invisible cell phenotypes with conditional generative modeling
Biological research relies on observing cell phenotypes, often obscured by natural variability. Here, the authors use generative modelling to unveil hidden changes triggered by infections, mutations, or drugs, allowing for accessible discovery of biomarkers.
- Alexis Lamiable
- , Tiphaine Champetier
- & Auguste Genovesio
-
Article
| Open AccessTripartite motif containing 26 prevents steatohepatitis progression by suppressing C/EBPδ signalling activation
Nonalcoholic steatohepatitis (NASH) is a heterogeneous disease with complicated pathogenesis. Here the authors identify that the E3 ligase TRIM26 confers protection against NASH development via suppression of CCAAT/enhancer binding protein delta (C/EBPδ).
- Minxuan Xu
- , Jun Tan
- & Lianyi Han
-
Article
| Open AccessProteomics of CKD progression in the chronic renal insufficiency cohort
Progression of chronic kidney disease may lead to kidney failure and cardiovascular, metabolic and bone disease complications. Here, the authors conduct a large-scale proteomic study in patients with chronic kidney disease, identify numerous proteins that predict kidney failure, some of which are likely causal mediators and hence potential therapeutic targets.
- Ruth F. Dubin
- , Rajat Deo
- & Peter Ganz
-
Article
| Open AccessMetabolic phenotyping of BMI to characterize cardiometabolic risk: evidence from large population-based cohorts
A substantial proportion of patients with cardiometabolic conditions appear to have normal BMI. Conversely, not all obese individuals exhibit these disorders. Here, we show that a metabolic BMI score derived from lipidomic models, allows quantify metabolic dysregulation in obesity independently of BMI.
- Habtamu B. Beyene
- , Corey Giles
- & Peter J. Meikle
-
Article
| Open AccessCombining Asian and European genome-wide association studies of colorectal cancer improves risk prediction across racial and ethnic populations
Most genetic studies have been done on European cohorts, which affects the efficacy of polygenic risk scores in non-European populations. Here, the authors demonstrate that a colorectal cancer PRS including Asian and European ancestries has improved performance over the European-centric PRS across racial and ethnic groups.
- Minta Thomas
- , Yu-Ru Su
- & Li Hsu
-
Article
| Open AccessMultimodal analysis of cell-free DNA whole-methylome sequencing for cancer detection and localization
cfDNA is a promising tool to detect early cancer, however, it has been limited by detection limits and technological ability. Here, the authors develop a method for simultaneous cfDNA methylation, fragmentation and copy number alteration sequencing and apply to a cohort of healthy and cancer patients.
- Fenglong Bie
- , Zhijie Wang
- & Shugeng Gao
-
Article
| Open AccessSocial and psychological adversity are associated with distinct mother and infant gut microbiome variations
Here, in a cohort of mother-child dyads, the authors show that maternal prenatal social disadvantage and psychosocial stressors associate with distinct gut microbiome taxonomic and functional diversity in both the mothers and their four-month children.
- Barbara B. Warner
- , Bruce A. Rosa
- & Makedonka Mitreva
-
Article
| Open AccessPlasma metabolomic profiles associated with mortality and longevity in a prospective analysis of 13,512 individuals
The metabolic alterations underpinning aging processes and mortality in humans are not well understood. Here, the authors show that differences in levels of nucleosides, amino acids, and several lipid subclasses can predict mortality and longevity.
- Fenglei Wang
- , Anne-Julie Tessier
- & Marta Guasch-Ferré
-
Article
| Open AccessDetection of diagnostic and prognostic methylation-based signatures in liquid biopsy specimens from patients with meningiomas
Recurrence of meningiomas is unpredictable by current methods based on surgically removed specimens, and identification of patients likely to recur could inform treatment strategy. Here, the authors analysed DNA methylation in liquid biopsy specimens from meningioma patients to help classify recurrence risk noninvasively even before surgery.
- Grayson A. Herrgott
- , James M. Snyder
- & Houtan Noushmehr
-
Article
| Open AccessCSF proteome profiling reveals biomarkers to discriminate dementia with Lewy bodies from Alzheimer´s disease
This study characterizes the CSF proteome changes underlying Dementia with Lewy Bodies (DLB) and identifies pathophysiological and diagnostic leads associated to this cause of dementia. Findings have been translated into a biomarker panel that could identify DLB patients with high accuracy across different cohorts.
- Marta del Campo
- , Lisa Vermunt
- & Charlotte E. Teunissen
-
Article
| Open AccessQuantifying the causal impact of biological risk factors on healthcare costs
Understanding the causal impact that risk factors have on healthcare cost is critical to evaluate healthcare interventions. Here, authors show that waist circumference, body mass index, and blood pressure have robust causal impact on healthcare cost.
- Jiwoo Lee
- , Sakari Jukarainen
- & Andrea Ganna
-
Article
| Open AccessExosomal circEZH2_005, an intestinal injury biomarker, alleviates intestinal ischemia/reperfusion injury by mediating Gprc5a signaling
Currently, intestinal ischemia/reperfusion injury lacks ideal diagnostic markers and effective therapeutic targets. Here, the authors show exosomal circEZH2_005 is an intestinal injury biomarker and protects the intestine from ischemia/reperfusion injury by promoting crypt cell proliferation via affecting hnRNPA1- mediated Gprc5a stability.
- Wenjuan Zhang
- , Bowei Zhou
- & Ke-Xuan Liu
-
Article
| Open AccessMulti-trait discovery and fine-mapping of lipid loci in 125,000 individuals of African ancestry
There have been few genome-wide association studies that analyze multiple lipid traits simultaneously, especially in individuals of African ancestry. To address this, the authors performed a multi-trait analysis of GWAS along with fine-mapping to find new genetic loci associated with lipid traits in individuals of African ancestry.
- Abram Bunya Kamiza
- , Sounkou M. Touré
- & Segun Fatumo
-
Article
| Open AccessThe Oncology Biomarker Discovery framework reveals cetuximab and bevacizumab response patterns in metastatic colorectal cancer
Identifying actionable biomarkers remains a challenge. Here, the authors develop a framework Oncology Biomarker Discovery (OncoBird), apply it to a phase III trial and investigate the molecular and biomarker landscape of metastatic colorectal carcinoma patients.
- Alexander J. Ohnmacht
- , Arndt Stahler
- & Michael P. Menden
-
Article
| Open AccessPortosystemic shunt placement reveals blood signatures for the development of hepatic encephalopathy through mass spectrometry
Patients with liver disease undergoing transjugular intrahepatic portosystemic shunt (TIPS) are at a higher risk of hepatic encephalopathy (HE). Here, the authors show intrahepatic shunting and specific metabolites, especially bile acids, as potential biomarkers and treatment targets for HE.
- Ana Carolina Dantas Machado
- , Stephany Flores Ramos
- & Amir Zarrinpar
-
Article
| Open AccessPractical diagnosis of cirrhosis in non-alcoholic fatty liver disease using currently available non-invasive fibrosis tests
The practical rules for the early non-invasive diagnosis of cirrhosis in NAFLD are not well defined. Here, the authors develop and validate two diagnostic tools: a stepwise stratification algorithm including a cirrhosis group, and a risk prediction chart providing a personalized assessment of the individual probability of cirrhosis.
- Jérôme Boursier
- , Marine Roux
- & Philip N. Newsome
-
Article
| Open AccessA versatile, fast and unbiased method for estimation of gene-by-environment interaction effects on biobank-scale datasets
Identification of gene-by-environment interactions is crucial to understand the interplay of environmental effects on complex traits. Here, the authors present MonsterLM, a method for estimating the proportion of trait variance explained by gene-by-environment interactions in a fast, unbiased manner on biobank-scale datasets.
- Matteo Di Scipio
- , Mohammad Khan
- & Guillaume Paré
-
Article
| Open AccessGut microbiota and fecal short chain fatty acids differ with adiposity and country of origin: the METS-microbiome study
Here, using amplicon sequencing and metabolomics in a large multi-country cohort, the authors find that adiposity-related microbiota differences differ between low-to-middle-income compared to high-income countries.
- Gertrude Ecklu-Mensah
- , Candice Choo-Kang
- & Lara R. Dugas
-
Article
| Open AccessDiurnal rhythms of wrist temperature are associated with future disease risk in the UK Biobank
Many chronic diseases present with desynchronized sleep-wake cycles, indicative of disrupted biorhythms. Here, the authors propose peripheral thermoregulation as a digital biomarker based on the association between lower temperature rhythms measured from wearable activity trackers with future onset of disease, such as type 2 diabetes, non-alcoholic fatty liver disease, hypertension and pneumonia.
- Thomas G. Brooks
- , Nicholas F. Lahens
- & Carsten Skarke
-
Article
| Open AccessBiology-guided deep learning predicts prognosis and cancer immunotherapy response
The clinical application of data-driven deep learning models remains challenging. Here, a biology-guided deep learning approach allows the simultaneous prediction of the tumour immune and stromal microenvironment status as well as treatment outcomes from medical images in gastric cancer.
- Yuming Jiang
- , Zhicheng Zhang
- & Ruijiang Li
-
Article
| Open AccessAt-home wearables and machine learning sensitively capture disease progression in amyotrophic lateral sclerosis
“In ALS clinical trials, efficacy is often assessed via subjective patient or clinician reports. The authors introduce a machine-learned severity score based on wearable sensor data from daily activities, which is reliable, sensitive, and could reduce clinical trial sizes.”
- Anoopum S. Gupta
- , Siddharth Patel
- & Fernando Vieira
-
Article
| Open AccessiCLOTS: open-source, artificial intelligence-enabled software for analyses of blood cells in microfluidic and microscopy-based assays
Microscopy has undoubtedly advanced biomedical research, but novel hypotheses are often lost to a lack of analytical tools. Here authors propose iCLOTS, a freely-available software that allows researchers to apply image processing and artificial intelligence algorithms to their own data.
- Meredith E. Fay
- , Oluwamayokun Oshinowo
- & Wilbur A. Lam
-
Article
| Open AccessElucidation of the low-expressing erythroid CR1 phenotype by bioinformatic mining of the GATA1-driven blood-group regulome
This study provides a systematic approach to explore how blood group expression is regulated by transcription factors. As proof-of-principle, the genetic basis underlying the very low levels of CR1 on red cells of the Helgeson phenotype is explained.
- Ping Chun Wu
- , Yan Quan Lee
- & Martin L. Olsson
-
Article
| Open AccessAssociation between vmPFC gray matter volume and smoking initiation in adolescents
The relationship between brain development and smoking behaviour is not well understood. Here, the authors show an association between volume of the left ventromedial prefrontal cortex and smoking initiation in adolescents.
- Shitong Xiang
- , Tianye Jia
- & Jianfeng Feng
-
Article
| Open AccessDeep-learning based detection of vessel occlusions on CT-angiography in patients with suspected acute ischemic stroke
AI may enhance diagnostic accuracy in medicine. Here, authors developed an AI model to detect and localise vessel occlusions in patients with suspected ischemic stroke, outperforming commercial tools on pseudo-prospective multicenter benchmarking.
- Gianluca Brugnara
- , Michael Baumgartner
- & Philipp Vollmuth
-
Article
| Open AccessDeep learning for obstructive sleep apnea diagnosis based on single channel oximetry
Obstructive sleep apnea (OSA) is a highly prevalent condition with a number of comorbidities, but it has a high misdiagnosis rate. In this retrospective study, the authors used six independent databases to develop and evaluate a deep learning model, denoted OxiNet, for the diagnosis of OSA from the sole oximetry signal.
- Jeremy Levy
- , Daniel Álvarez
- & Joachim A. Behar
-
Article
| Open AccessA machine-learning approach to human ex vivo lung perfusion predicts transplantation outcomes and promotes organ utilization
Ex vivo perfusion is a unique platform to study isolated human lungs. Here, authors show that a machine learning model, InsighTx, derived from data generated during ex vivo lung perfusion can accurately predict transplant outcomes and increase organ utilization rates.
- Andrew T. Sage
- , Laura L. Donahoe
- & Shaf Keshavjee
-
Article
| Open AccessSpatial transcriptomics analysis of esophageal squamous precancerous lesions and their progression to esophageal cancer
Understanding the molecular changes in the transition from esophageal squamous precancerous lesions (ESPL) to esophageal squamous cell carcinoma (ESCC) remains essential. Here, the authors analyze ESPL samples using spatial transcriptomics and reveal expression changes in TAGLN2 and CRNN during progression to ESCC.
- Xuejiao Liu
- , Simin Zhao
- & Zigang Dong
-
Article
| Open AccessMulti-PGS enhances polygenic prediction by combining 937 polygenic scores
Polygenic scores (PGS) have high potential for clinical use but are currently underpowered for many applications. Here, the authors develop an approach that leverages an agnostic library of hundreds of PGS to increase prediction of complex diseases and other traits. This multi-PGS framework is ideal for emerging biobank data.
- Clara Albiñana
- , Zhihong Zhu
- & Bjarni J. Vilhjálmsson
-
Article
| Open AccessMAPK inhibitor sensitivity scores predict sensitivity driven by the immune infiltration in pediatric low-grade gliomas
The MAPK pathway is a key driver of pediatric low-grade gliomas (pLGG); however, response to MAPK inhibitors (MAPKi) in pLGG patients is not consistent. Here, the authors develop MAPKi sensitivity scores (MSS) to predict response to MAPKi and apply them to bulk and single-cell sequencing datasets from pLGG patients and preclinical models.
- Romain Sigaud
- , Thomas K. Albert
- & Till Milde
-
Article
| Open AccessIncreasing number of long-lived ancestors marks a decade of healthspan extension and healthier metabolomics profiles
Although human life expectancy has been increasing, time spent in good physical and cognitive health has not been rising at similar rate. Here, the authors show that both lifespan and healthspan are quantitatively linked to ancestral longevity, and that those from the longest-lived families have a healthier metabolomics profile before the onset of disease, highlighting the important role of the family in healthy survival.
- Niels van den Berg
- , Mar Rodríguez-Girondo
- & Marian Beekman
-
Article
| Open AccessClearance of defective muscle stem cells by senolytics restores myogenesis in myotonic dystrophy type 1
Muscle stem cells drive muscle regeneration and are affected in myotonic dystrophy type 1. Here, the authors demonstrate that some muscle stem cells show signs of senescence in myotonic dystrophy type 1 and administer senolytics to eliminate these defective cells and restore myogenesis.
- Talita C. Conte
- , Gilberto Duran-Bishop
- & Nicolas A. Dumont
-
Article
| Open AccessNext generation pan-cancer blood proteome profiling using proximity extension assay
Comprehensive and scalable proteomic profiling of plasma samples can improve the screening and diagnosis of cancer patients. Here, the authors use the Olink Proximity Extension Assay technology to characterise the plasma proteomes of 1477 patients across twelve cancer types, and use machine learning to obtain a protein panel for cancer classification.
- María Bueno Álvez
- , Fredrik Edfors
- & Mathias Uhlén
-
Article
| Open AccessExtending the dynamic range of biomarker quantification through molecular equalization
The human plasma proteome spans over 10 orders of magnitude, but current quantification methods cover only 3-4 orders. Here, authors propose a tuning approach for proximity-based assays, enabling simultaneous quantification of four analytes across a wide concentration range (low fM to mid-nM).
- Sharon S. Newman
- , Brandon D. Wilson
- & H. Tom Soh
-
Article
| Open AccessOpportunistic detection of type 2 diabetes using deep learning from frontal chest radiographs
Traditional type 2 diabetes (T2D) screening methods often depend on age, BMI guidelines and glucose measurements. Here, authors use a deep learning model that leverages chest radiographs and electronic health record data to screen for T2D, highlighting potential for early detection and intervention.
- Ayis Pyrros
- , Stephen M. Borstelmann
- & William Galanter
-
Article
| Open AccessOptimal strategies for learning multi-ancestry polygenic scores vary across traits
There are challenges with transferring genetic risk scores from ancestry in which they were generated to another. Here, the authors investigate the use of multi-ancestry versus single-ancestry training sets to construct polygenic scores and find that the optimal strategy varies across traits.
- Brieuc Lehmann
- , Maxine Mackintosh
- & Chris Holmes
-
Article
| Open AccessSUMOylation of Rho-associated protein kinase 2 induces goblet cell metaplasia in allergic airways
Allergic asthma is characterized by goblet cell metaplasia. Here, the authors show protein SUMOylation contributes to goblet cell metaplasia and SUMOylation-mediated ROCK2 activation is an integral component of Rho/ROCK signalling pathway in controlling the airway goblet cell metaplasia.
- Dan Tan
- , Meiping Lu
- & Ximei Wu
-
Article
| Open AccessThe transcriptional landscape and diagnostic potential of long non-coding RNAs in esophageal squamous cell carcinoma
Esophageal squamous cell carcinoma is difficult to detect at early stages, and late detection is often linked to poor prognosis. Here, the authors develop a lncRNA signature predictive of ESCC and validate across multiple external cohorts.
- Meng Zhou
- , Siqi Bao
- & Zhihua Liu
-
Article
| Open AccessNivolumab plus chemoradiotherapy in locally-advanced cervical cancer: the NICOL phase 1 trial
A combination of chemoradiotherapy followed by brachytherapy is recommended for patients with locally-advanced cervical cancer (LACC), however there is still a high risk of disease recurrence. Here the authors report clinical outcomes and immunologic correlates of a clinical trial of the PD-1 inhibitor nivolumab in combination with chemoradiotherapy in LACC patients.
- Manuel Rodrigues
- , Giulia Vanoni
- & Emanuela Romano
-
Article
| Open AccessUBAP2 plays a role in bone homeostasis through the regulation of osteoblastogenesis and osteoclastogenesis
The authors identify UBAP2 as a novel osteoporosis susceptibility gene by performing association studies focusing on coding regions of the genome, and report that it plays a role in bone homeostasis through the regulation of bone remodelling.
- Jeonghyun Kim
- , Bo-Young Kim
- & Seon-Yong Jeong
-
Article
| Open AccessA dorsomedial prefrontal cortex-based dynamic functional connectivity model of rumination
Rumination, which is the tendency to dwell on negative internal states repetitively, is a well-known cognitive style associated with depression. The authors developed a predictive model of rumination and observed that the dorsomedial prefrontal cortex plays an important role in rumination.
- Jungwoo Kim
- , Jessica R. Andrews-Hanna
- & Choong-Wan Woo
-
Article
| Open AccessIdentification of BRCA1/2 mutation female carriers using circulating microRNA profiles
BRCA1/2 mutations are known to increase risk of breast and ovarian cancer but carrier status in healthy individuals is unknown without genetic testing. Here, the authors created a circulating miRNA signature to predict BRCA1/2 carrier status in healthy individuals to aid the decision process on genetic testing.
- Kevin Elias
- , Urszula Smyczynska
- & Dipanjan Chowdhury