Search

Abstract Patients with high-grade glioma (HGG) have an extremely poor prognosis compounded by a lack of advancement in clinical care over the past few decades. Regardless of classification, most newly diagnosed patients receive the same treatment, radiation and temozolomide (RT/TMZ). We developed a functional precision oncology test that prospectively identifies individual patient’s response to this treatment regimen. Tumor tissues isolated from patients with newly diagnosed HGG enrolled in 3D PREDICT REGISTRY were evaluated for response to chemotherapeutic agents using the 3D Predict™ Glioma test. Patients receiving RT/TMZ were followed for 2 years. Clinical outcomes including imaging, assessments, and biomarker measurements were compared to patient matched test-predicted therapy response. Median survival between test-predicted temozolomide responders and test-predicted temozolomide non-responders revealed a statistically significant increase in progression-free survival when using the test to predict response across multiple subgroups including HGG (5.8 months), glioblastoma (4.7 months), and MGMT unmethylated glioblastoma (4.7 months). Overall survival was also positively separated across the subgroups at 7.6, 5.1, and 6.3 months respectively. The strong correlation of 3D Predict Glioma test results with clinical outcomes demonstrates that this functional test is prognostic in patients treated with RT/TMZ and supports aligning clinical treatment to test-predicted response across varying HGG subgroups.

Abstract To survive extreme desiccation, seeds enter a period of quiescence that can last millennia. Seed quiescence involves the accumulation of protective storage proteins and lipids through unknown adjustments in protein homeostasis (proteostasis). Here, we show that mutation of all six type–II metacaspase (MCA–II) proteases in Arabidopsis thaliana disturbs proteostasis in seeds. MCA–II mutant seeds fail to restrict the AAA ATPase CELL DIVISION CYCLE 48 (CDC48) at the endoplasmic reticulum to discard misfolded proteins, compromising seed storability. Endoplasmic reticulum (ER) localization of CDC48 relies on the MCA–IIs-dependent cleavage of PUX10 (ubiquitination regulatory X domain–containing 10), the adaptor protein responsible for titrating CDC48 to lipid droplets. PUX10 cleavage enables the shuttling of CDC48 between lipid droplets and the ER, providing an important regulatory mechanism sustaining spatiotemporal proteolysis, lipid droplet dynamics, and protein homeostasis. In turn, the removal of the PUX10 adaptor in MCA–II mutant seeds partially restores proteostasis, CDC48 localization, and lipid droplet dynamics prolonging seed lifespan. Taken together, we uncover a proteolytic module conferring seed longevity.

Abstract Through technological innovations, patient cohorts can be examined from multiple views with high-dimensional, multiscale biomedical data to classify clinical phenotypes and predict outcomes. Here, we aim to present our approach for analyzing multimodal data using unsupervised and supervised sparse linear methods in a COVID-19 patient cohort. This prospective cohort study of 149 adult patients was conducted in a tertiary care academic center. First, we used sparse canonical correlation analysis (CCA) to identify and quantify relationships across different data modalities, including viral genome sequencing, imaging, clinical data, and laboratory results. Then, we used cooperative learning to predict the clinical outcome of COVID-19 patients: Intensive care unit admission. We show that serum biomarkers representing severe disease and acute phase response correlate with original and wavelet radiomics features in the LLL frequency channel (cor(Xu 1, Zv 1) = 0.596, p value

Abstract Generative Artificial Intelligence is set to revolutionize healthcare delivery by transforming traditional patient care into a more personalized, efficient, and proactive process. Chatbots, serving as interactive conversational models, will probably drive this patient-centered transformation in healthcare. Through the provision of various services, including diagnosis, personalized lifestyle recommendations, dynamic scheduling of follow-ups, and mental health support, the objective is to substantially augment patient health outcomes, all the while mitigating the workload burden on healthcare providers. The life-critical nature of healthcare applications necessitates establishing a unified and comprehensive set of evaluation metrics for conversational models. Existing evaluation metrics proposed for various generic large language models (LLMs) demonstrate a lack of comprehension regarding medical and health concepts and their significance in promoting patients’ well-being. Moreover, these metrics neglect pivotal user-centered aspects, including trust-building, ethics, personalization, empathy, user comprehension, and emotional support. The purpose of this paper is to explore state-of-the-art LLM-based evaluation metrics that are specifically applicable to the assessment of interactive conversational models in healthcare. Subsequently, we present a comprehensive set of evaluation metrics designed to thoroughly assess the performance of healthcare chatbots from an end-user perspective. These metrics encompass an evaluation of language processing abilities, impact on real-world clinical tasks, and effectiveness in user-interactive conversations. Finally, we engage in a discussion concerning the challenges associated with defining and implementing these metrics, with particular emphasis on confounding factors such as the target audience, evaluation methods, and prompt techniques involved in the evaluation process.

Abstract Transglutaminase 1 (TGM1) plays an essential role in skin barrier formation by cross-linking proteins in differentiated keratinocytes. Here, we established a protocol for the antibody-dependent detection of TGM1 protein and the parallel detection of TGM activity. TGM1 immunoreactivity initially increased and co-localized with membrane-associated TGM activity during keratinocyte differentiation. TGM activity persisted upon further differentiation of keratinocytes, whereas TGM1 immunoreactivity was lost under standard assay conditions. Pretreatment of tissue sections with the proteases trypsin or proteinase K enabled immunodetection of TGM1 in cornified keratinocytes, indicating that removal of other proteins was a prerequisite for TGM1 immunolabeling after cornification. The increase of TGM activity and subsequent loss of TGM1 immunoreactivity could be replicated in HEK293T cells transfected with TGM1, suggesting that protein cross-linking mediated by TGM1 itself may lead to reduced recognition of TGM1 by antibodies. To screen for proteins potentially regulating TGM1, we performed Virotrap experiments and identified the CAPNS1 subunit of calpain as an interaction partner of TGM1. Treatment of keratinocytes and TGM1-transfected HEK293T cells with chemical inhibitors of calpain suppressed transglutamination. Our findings suggest that calpain contributes to the control of TGM1-mediated transglutamination and proteins cross-linked by transglutamination mask epitopes of TGM1.

Abstract Patients experiencing mental health crises often seek help through messaging-based platforms, but may face long wait times due to limited message triage capacity. Here we build and deploy a machine-learning-enabled system to improve response times to crisis messages in a large, national telehealth provider network. We train a two-stage natural language processing (NLP) system with key word filtering followed by logistic regression on 721 electronic medical record chat messages, of which 32% are potential crises (suicidal/homicidal ideation, domestic violence, or non-suicidal self-injury). Model performance is evaluated on a retrospective test set (4/1/21–4/1/22, N = 481) and a prospective test set (10/1/22–10/31/22, N = 102,471). In the retrospective test set, the model has an AUC of 0.82 (95% CI: 0.78–0.86), sensitivity of 0.99 (95% CI: 0.96–1.00), and PPV of 0.35 (95% CI: 0.309–0.4). In the prospective test set, the model has an AUC of 0.98 (95% CI: 0.966–0.984), sensitivity of 0.98 (95% CI: 0.96–0.99), and PPV of 0.66 (95% CI: 0.626–0.692). The daily median time from message receipt to crisis specialist triage ranges from 8 to 13 min, compared to 9 h before the deployment of the system. We demonstrate that a NLP-based machine learning model can reliably identify potential crisis chat messages in a telehealth setting. Our system integrates into existing clinical workflows, suggesting that with appropriate training, humans can successfully leverage ML systems to facilitate triage of crisis messages.

Abstract Public proteomics data often lack essential metadata, limiting its potential. To address this, we present lesSDRF, a tool to simplify the process of metadata annotation, thereby ensuring that data leave a lasting, impactful legacy well beyond its initial publication.

Abstract Most eukaryotic proteins are N-terminally acetylated, but the functional impact on a global scale has remained obscure. Using genome-wide CRISPR knockout screens in human cells, we reveal a strong genetic dependency between a major N-terminal acetyltransferase and specific ubiquitin ligases. Biochemical analyses uncover that both the ubiquitin ligase complex UBR4-KCMF1 and the acetyltransferase NatC recognize proteins bearing an unacetylated N-terminal methionine followed by a hydrophobic residue. NatC KO-induced protein degradation and phenotypes are reversed by UBR knockdown, demonstrating the central cellular role of this interplay. We reveal that loss of Drosophila NatC is associated with male sterility, reduced longevity, and age-dependent loss of motility due to developmental muscle defects. Remarkably, muscle-specific overexpression of UbcE2M, one of the proteins targeted for NatC KO-mediated degradation, suppresses defects of NatC deletion. In conclusion, NatC-mediated N-terminal acetylation acts as a protective mechanism against protein degradation, which is relevant for increased longevity and motility.

Abstract Intra-tumoral heterogeneity and cell-state plasticity are key drivers for the therapeutic resistance of glioblastoma. Here, we investigate the association between spatial cellular organization and glioblastoma prognosis. Leveraging single-cell RNA-seq and spatial transcriptomics data, we develop a deep learning model to predict transcriptional subtypes of glioblastoma cells from histology images. Employing this model, we phenotypically analyze 40 million tissue spots from 410 patients and identify consistent associations between tumor architecture and prognosis across two independent cohorts. Patients with poor prognosis exhibit higher proportions of tumor cells expressing a hypoxia-induced transcriptional program. Furthermore, a clustering pattern of astrocyte-like tumor cells is associated with worse prognosis, while dispersion and connection of the astrocytes with other transcriptional subtypes correlate with decreased risk. To validate these results, we develop a separate deep learning model that utilizes histology images to predict prognosis. Applying this model to spatial transcriptomics data reveal survival-associated regional gene expression programs. Overall, our study presents a scalable approach to unravel the transcriptional heterogeneity of glioblastoma and establishes a critical connection between spatial cellular architecture and clinical outcomes.

Abstract We propose an interpretable and scalable model to predict likely diagnoses at an encounter based on past diagnoses and lab results. This model is intended to aid physicians in their interaction with the electronic health records (EHR). To accomplish this, we retrospectively collected and de-identified EHR data of 2,701,522 patients at Stanford Healthcare over a time period from January 2008 to December 2016. A population-based sample of patients comprising 524,198 individuals (44% M, 56% F) with multiple encounters with at least one frequently occurring diagnosis codes were chosen. A calibrated model was developed to predict ICD-10 diagnosis codes at an encounter based on the past diagnoses and lab results, using a binary relevance based multi-label modeling strategy. Logistic regression and random forests were tested as the base classifier, and several time windows were tested for aggregating the past diagnoses and labs. This modeling approach was compared to a recurrent neural network based deep learning method. The best model used random forest as the base classifier and integrated demographic features, diagnosis codes, and lab results. The best model was calibrated and its performance was comparable or better than existing methods in terms of various metrics, including a median AUROC of 0.904 (IQR [0.838, 0.954]) over 583 diseases. When predicting the first occurrence of a disease label for a patient, the median AUROC with the best model was 0.796 (IQR [0.737, 0.868]). Our modeling approach performed comparably as the tested deep learning method, outperforming it in terms of AUROC (p

Abstract Addressing the elusive specificity of cysteine cathepsins, which in contrast to caspases and trypsin-like proteases lack strict specificity determining P1 pocket, calls for innovative approaches. Proteomic analysis of cell lysates with human cathepsins K, V, B, L, S, and F identified 30,000 cleavage sites, which we analyzed by software platform SAPS-ESI (Statistical Approach to Peptidyl Substrate-Enzyme Specific Interactions). SAPS-ESI is used to generate clusters and training sets for support vector machine learning. Cleavage site predictions on the SARS-CoV-2 S protein, confirmed experimentally, expose the most probable first cut under physiological conditions and suggested furin-like behavior of cathepsins. Crystal structure analysis of representative peptides in complex with cathepsin V reveals rigid and flexible sites consistent with analysis of proteomics data by SAPS-ESI that correspond to positions with heterogeneous and homogeneous distribution of residues. Thereby support for design of selective cleavable linkers of drug conjugates and drug discovery studies is provided.

Abstract The success of training computer-vision models heavily relies on the support of large-scale, real-world images with annotations. Yet such an annotation-ready dataset is difficult to curate in pathology due to the privacy protection and excessive annotation burden. To aid in computational pathology, synthetic data generation, curation, and annotation present a cost-effective means to quickly enable data diversity that is required to boost model performance at different stages. In this study, we introduce a large-scale synthetic pathological image dataset paired with the annotation for nuclei semantic segmentation, termed as Synthetic Nuclei and annOtation Wizard (SNOW). The proposed SNOW is developed via a standardized workflow by applying the off-the-shelf image generator and nuclei annotator. The dataset contains overall 20k image tiles and 1,448,522 annotated nuclei with the CC-BY license. We show that SNOW can be used in both supervised and semi-supervised training scenarios. Extensive results suggest that synthetic-data-trained models are competitive under a variety of model training settings, expanding the scope of better using synthetic images for enhancing downstream data-driven clinical tasks.

Steyaert, Qiu et al. develop a deep learning framework for multimodal data fusion for adult and pediatric brain tumors. Multimodal data models combining histopathology imaging and gene expression data outperform single data models in predicting prognosis.

Abstract Heparan sulfates (HS) proteoglycans are commonly found on the cell surface and mediate many processes. Binding of HS ligands is determined by the sulfation code on the HS chain that can be N-/2-O/6-O- or 3-O-sulfated, generating heterogenous sulfation patterns. 3-O sulfated HS (3S-HS) play a role in several (patho)physiological processes such as blood coagulation, viral pathogenesis and binding and internalization of tau in Alzheimer’s disease. However, few 3S-HS-specific interactors are known. Thus, our insight into the role of 3S-HS in health and disease is limited, especially in the central nervous system. Using human CSF, we determined the interactome of synthetic HS with defined sulfation patterns. Our affinity-enrichment mass spectrometry studies expand the repertoire of proteins that may interact with (3S-)HS. Validating our approach, ATIII, a known 3S-HS interactor, was found to require GlcA-GlcNS6S3S for binding, similar to what has been reported. Our dataset holds novel, potential HS and 3S-HS protein ligands, that can be explored in future studies focusing on molecular mechanisms that depend on 3S-HS in (patho)physiological conditions.

Feeding an increasing world population in the context of climate change is one of the grand challenges faced by our generation. Here, the authors show that external application of synthetic complementary peptides can increase the abundance of target proteins to modulate plant growth or stress resistance.

Abstract Lack of accurate methods for early lymphoma detection limits the ability to cure patients. Since patients with Non-Hodgkin lymphomas (NHL) who present with advanced disease have worse outcomes, accurate and sensitive methods for early detection are needed to improve patient care. We developed a DNA methylation-based prediction tool for NHL, based on blood samples collected prospectively from 278 apparently healthy patients who were followed for up to 16 years to monitor for NHL development. A predictive score was developed using machine learning methods in a robust training/validation framework. Our predictive score incorporates CpG DNA methylation at 135 genomic positions, with higher scores predicting higher risk. It was 85% and 78% accurate for identifying patients at risk of developing future NHL, in patients with high or low epigenetic mitotic clock respectively, in a validation cohort. It was also sensitive at detecting active NHL (96.3% accuracy) and healthy status (95.6% accuracy) in additional independent cohorts. Scores optimized for specific NHL subtypes showed significant but lower accuracy for predicting other subtypes. Our score incorporates hyper-methylation of Polycomb and HOX genes, which have roles in NHL development, as well as PAX5 - a master transcriptional regulator of B-cell fate. Subjects with higher risk scores showed higher regulatory T-cells, memory B-cells, but lower naïve T helper lymphocytes fractions in the blood. Future prospective studies will be required to confirm the utility of our signature for managing patients who are at high risk for developing future NHL.

Abstract Developing prediction models for emerging infectious diseases from relatively small numbers of cases is a critical need for improving pandemic preparedness. Using COVID-19 as an exemplar, we propose a transfer learning methodology for developing predictive models from multi-modal electronic healthcare records by leveraging information from more prevalent diseases with shared clinical characteristics. Our novel hierarchical, multi-modal model ( $${\textsc {TransMED}}$$ T R A N S MED ) integrates baseline risk factors from the natural language processing of clinical notes at admission, time-series measurements of biomarkers obtained from laboratory tests, and discrete diagnostic, procedure and drug codes. We demonstrate the alignment of $${\textsc {TransMED}}$$ T R A N S MED ’s predictions with well-established clinical knowledge about COVID-19 through univariate and multivariate risk factor driven sub-cohort analysis. $${\textsc {TransMED}}$$ T R A N S MED ’s superior performance over state-of-the-art methods shows that leveraging patient data across modalities and transferring prior knowledge from similar disorders is critical for accurate prediction of patient outcomes, and this approach may serve as an important tool in the early response to future pandemics.

Abstract Breast carcinoma is the most common cancer among women worldwide that consists of a heterogeneous group of subtype diseases. The whole-slide images (WSIs) can capture the cell-level heterogeneity, and are routinely used for cancer diagnosis by pathologists. However, key driver genetic mutations related to targeted therapies are identified by genomic analysis like high-throughput molecular profiling. In this study, we develop a deep-learning model to predict the genetic mutations and biological pathway activities directly from WSIs. Our study offers unique insights into WSI visual interactions between mutation and its related pathway, enabling a head-to-head comparison to reinforce our major findings. Using the histopathology images from the Genomic Data Commons Database, our model can predict the point mutations of six important genes (AUC 0.68–0.85) and copy number alteration of another six genes (AUC 0.69–0.79). Additionally, the trained models can predict the activities of three out of ten canonical pathways (AUC 0.65–0.79). Next, we visualized the weight maps of tumor tiles in WSI to understand the decision-making process of deep-learning models via a self-attention mechanism. We further validated our models on liver and lung cancers that are related to metastatic breast cancer. Our results provide insights into the association between pathological image features, molecular outcomes, and targeted therapies for breast cancer patients.

Plants respond to warmth via growth processes termed thermomorphogenesis. Here, via a phosphoproteomics approach, the authors show that the mitogen activated protein kinase TOT3 regulates thermomorphogenesis in both wheat and Arabidopsis and modifies brassinosteroid signaling in Arabidopsis.

Abstract The COVID-19 pandemic overwhelms the medical resources in the stressed intensive care unit (ICU) capacity and the shortage of mechanical ventilation (MV). We performed CT-based analysis combined with electronic health records and clinical laboratory results on Cohort 1 (n = 1662 from 17 hospitals) with prognostic estimation for the rapid stratification of PCR confirmed COVID-19 patients. These models, validated on Cohort 2 (n = 700) and Cohort 3 (n = 662) constructed from nine external hospitals, achieved satisfying performance for predicting ICU, MV, and death of COVID-19 patients (AUROC 0.916, 0.919, and 0.853), even on events happened two days later after admission (AUROC 0.919, 0.943, and 0.856). Both clinical and image features showed complementary roles in prediction and provided accurate estimates to the time of progression (p

RNA-sequencing data from tumours can be used to predict the prognosis of patients. Here, the authors show that a neural network meta-learning approach can be useful for predicting prognosis from a small number of samples.

Realistic simulations of neurons and neural networks are key for understanding neural computations. Here the authors describe Neuron_Reduce, an analytic approach to simplify neurons receiving thousands of synapses and accelerate their simulations by 40–250 folds, while preserving voltage dynamics and dendritic computations.

A combination of large-scale connectomics with cellular and synapse data to generate a first draft statistical model of the neuron-to-neuron micro-connectome of a whole mouse neocortex. This micro-connectome recreates biological trends of targeting on the macro-, meso-, and micro-scale.

There is no universal reference material to develop extracellular vesicle (EV) separation methods and carry out calibration and normalization. Here the authors use HIV-derived gag proteins to assemble recombinant fluorescent EV as a trackable reference material resembling the physical and biochemical properties of sample EV.

Abstract Chromatin modifying enzymes are frequently mutated in cancer, resulting in widespread epigenetic deregulation. Recent reports indicate that inactivating mutations in the histone methyltransferase NSD1 define an intrinsic subtype of head and neck squamous cell carcinoma (HNSC) that features pronounced DNA hypomethylation. Here, we describe a similar hypomethylated subtype of lung squamous cell carcinoma (LUSC) that is enriched for both inactivating mutations and deletions in NSD1. The ‘NSD1 subtypes’ of HNSC and LUSC are highly correlated at the DNA methylation and gene expression levels, featuring ectopic expression of developmental transcription factors and genes that are also hypomethylated in Sotos syndrome, a congenital disorder caused by germline NSD1 mutations. Further, the NSD1 subtype of HNSC displays an ‘immune cold’ phenotype characterized by low infiltration of tumor-associated leukocytes, particularly macrophages and CD8+ T cells, as well as low expression of genes encoding the immunotherapy target PD-1 immune checkpoint receptor and its ligands. Using an in vivo model, we demonstrate that NSD1 inactivation results in reduced T cell infiltration into the tumor microenvironment, implicating NSD1 as a tumor cell-intrinsic driver of an immune cold phenotype. NSD1 inactivation therefore causes epigenetic deregulation across cancer sites, and has implications for immunotherapy.

Abstract The multi-domain transcriptional coactivators CBP/p300 integrate a multitude of signaling inputs, interacting with more than 400 proteins via one or more of their globular domains. While CBP/p300 function is typically considered in terms of these structured domains, about half of the protein consists of intrinsically disordered regions (IDRs) of varying length. However, these IDRs have only been thought of as linkers that allow flexible spatial arrangement of the structured domains, but recent studies have shown that similar IDRs mediate specific and critical interactions in other proteins. To examine the roles of IDRs in CBP, we performed yeast-two-hybrid screenings of placenta and lung cancer cDNA libraries, which demonstrated that the long IDR linking the KIX domain and bromodomain of CBP (termed ID3) can potentially bind to several proteins. The RNA-binding Zinc-finger protein 106 (ZFP106) detected in both libraries was identified as a novel substrate for CBP-mediated acetylation. Nuclear magnetic resonance (NMR) spectroscopy combined with cross-linking experiments and competition-binding assays showed that the fully disordered isolated ID3 transiently interacts with an IDR of ZFP106 in a fashion that disorder of both regions is maintained. These findings demonstrate that beside the linking function, ID3 can also interact with acetylation substrates of CBP.

Abstract Identification and validation of extracellular vesicle (EV)-associated biomarkers requires robust isolation and characterization protocols. We assessed the impact of some commonly implemented pre-analytical, analytical and post-analytical variables in EV research. Centrifugal filters with different membrane types and pore sizes are used to reduce large volume biofluids prior to EV isolation or to concentrate EVs. We compared five commonly reported filters for their efficiency when using plasma, urine and EV-spiked PBS. Regenerated cellulose membranes with pore size of 10 kDa recovered EVs the most efficient. Less than 40% recovery was achieved with other filters. Next, we analyzed the effect of the type of protein assays to measure EV protein in colorimetric and fluorometric kits. The fluorometric assay Qubit measured low concentration EV and BSA samples the most accurately with the lowest variation among technical and biological replicates. Lastly, we quantified Optiprep remnants in EV samples from density gradient ultracentrifugation and demonstrate that size-exclusion chromatography efficiently removes Optiprep from EVs. In conclusion, choice of centrifugal filters and protein assays confound EV analysis and should be carefully considered to increase efficiency towards biomarker discovery. SEC-based removal of Optiprep remnants from EVs can be considered for downstream applications.

The plant circadian clock regulates numerous developmental processes such as seasonal growth and flowering time. Here Ritteret al. identify two transcription factors, FRS7 and FRS12, which interact to form a repressor complex that regulates clock output partially by controlling the expression of GIGANTEA and PIF4.

A large portion of the proteome carries out its cellular function as part of macromolecular complexes. Here the authors describe Virotrap, a novel lysis-free approach for the isolation and identification of biologically relevant protein-protein and small molecule-protein interactions.

Nature Communications 8: Article number: 15235 (2017); Published online: 11 May 2017; Updated: 10 April 2018 The financial support for this Article was not correctly acknowledged. The Acknowledgements incorrectly stated that the study was funded in part by Research Foundation Flanders for project GN005212.