The multimodality cell segmentation challenge: toward universal solutions.

Cell segmentation is a critical step for quantitative single-cell analysis in microscopy images. Existing cell segmentation methods are often tailored to specific modalities or require manual interventions to specify hyper-parameters in different experimental settings. Here, we present a multimodality cell segmentation benchmark, comprising more than 1,500 labeled images derived from more than 50 diverse biological experiments. The top participants developed a Transformer-based deep-learning algorithm that not only exceeds existing methods but can also be applied to diverse microscopy images across imaging platforms and tissue types without manual parameter adjustments. This benchmark and the improved algorithm offer promising avenues for more accurate and versatile cell analysis in microscopy imaging.

A signature of platelet reactivity in CBC scattergrams reveals genetic predictors of thrombotic disease risk.

Genetic studies of platelet reactivity (PR) phenotypes may identify novel antiplatelet drug targets. However, such studies have been limited by small sample sizes (n < 5000) because of the complexity of measuring PR. We trained a model to predict PR from complete blood count (CBC) scattergrams. A genome-wide association study of this phenotype in 29 806 blood donors identified 21 distinct associations implicating 20 genes, of which 6 have been identified previously. The effect size estimates were significantly correlated with estimates from a study of flow cytometry-measured PR and a study of a phenotype of in vitro thrombus formation. A genetic score of PR built from the 21 variants was associated with the incidence rates of myocardial infarction and pulmonary embolism. Mendelian randomization analyses showed that PR was causally associated with the risks of coronary artery disease, stroke, and venous thromboembolism. Our approach provides a blueprint for using phenotype imputation to study the determinants of hard-to-measure but biologically important hematological traits.

A comparison of platelet function in cold-stored whole blood and platelet concentrates.

BACKGROUND: There is renewed interest in the use of whole blood (WB) for the resuscitation of trauma patients. Platelet function in stored WB compared to platelet concentrates is not well established and was assessed in vitro in this study. METHODS: Leucocyte-depleted cold-stored WB (CS-WB) was prepared using a Terumo WB-SP Imuflex kit and held at 2-6°C alongside: (A) UK standard pooled platelets stored at 20-24°C (RT-PLTS), (B) pooled platelets stored at 2-6°C (CS-PLTS), and (C) platelet-rich plasma produced using the Terumo kit (CS-PRP), for 21 days. A series of in vitro assays were assessed platelet function. RESULTS: Platelet count was retained to 57 ± 14% of starting number at day 21 in CS-WB. Over time, CS-WB platelets become more activated, with increased CD62P expression (day 1: 7 ± 3.7% vs. day 21: 59 ± 17.1%) and annexin V binding (day 1: 2 ± 0.2% vs. day 21: 21 ± 15.1%). For comparison, 18.6 ± 6% of platelets in RT-PLTS demonstrated CD62P expression at day 7, whereas annexin V binding in RT-PLTS at day 7 was 2.6 ± 0.5%. Over storage, aggregatory response to agonists decreased in all arms. Functional platelet microparticles increased steadily in CS-WB throughout storage. CONCLUSION: During storage, platelet count reduced in CS-WB, whereas CD62P expression and annexin V binding increased. This was accompanied by a reduced aggregatory response, although compared to 7-day-old RT-PLTS, CS-WB maintained a maximal response to agonists for longer, suggesting that the shelf life for CS-WB can be considered for up to 21 days.

High-throughput elucidation of thrombus formation reveals sources of platelet function variability.

In combination with microspotting, whole-blood microfluidics can provide high-throughput information on multiple platelet functions in thrombus formation. Based on assessment of the inter- and intra-subject variability in parameters of microspot-based thrombus formation, we aimed to determine the platelet factors contributing to this variation. Blood samples from 94 genotyped healthy subjects were analyzed for conventional platelet phenotyping: i.e. hematologic parameters, platelet glycoprotein (GP) expression levels and activation markers (24 parameters). Furthermore, platelets were activated by ADP, CRP-XL or TRAP. Parallel samples were investigated for whole-blood thrombus formation (6 microspots, providing 48 parameters of adhesion, aggregation and activation). Microspots triggered platelet activation through GP Ib-V-IX, GPVI, CLEC-2 and integrins. For most thrombus parameters, inter-subject variation was 2-4 times higher than the intra-subject variation. Principal component analyses indicated coherence between the majority of parameters for the GPVI-dependent microspots, partly linked to hematologic parameters, and glycoprotein expression levels. Prediction models identified parameters per microspot that were linked to variation in agonist-induced αIIbß3 activation and secretion. Common sequence variation of GP6 and FCER1G, associated with GPVI-induced αIIbß3 activation and secretion, affected parameters of GPVI-and CLEC-2-dependent thrombus formation. Subsequent analysis of blood samples from patients with Glanzmann thrombasthenia or storage pool disease revealed thrombus signatures of aggregation-dependent parameters that were subject-dependent, but not linked to GPVI activity. Taken together, this high-throughput elucidation of thrombus formation revealed patterns of inter-subject differences in platelet function, which were partly related to GPVI-induced activation and common genetic variance linked to GPVI, but also included a distinct platelet aggregation component.

Platelet Receptor Glycoprotein VI-Dimer Is Overexpressed in Patients with Atrial Fibrillation at High Risk of Ischemic Stroke.

Introduction Atrial fibrillation (AF) increases the risk of ischemic stroke (IS). We hypothesized that the functional form of platelet receptor glycoprotein (GP) VI, GPVI-dimer, which binds to collagen and fibrin causing platelet activation, is overexpressed in patients with AF who have not had a stroke. Methods A total of 75 inpatients with AF were recruited. None were admitted with or had previously had thrombotic events, including IS or myocardial infarction. Platelet surface expression of total GPVI, GPVI-dimer, and the platelet activation marker P-selectin were quantitated by whole blood flow cytometry. Serum biomarkers were collected in AF patients. Results were compared against patients contemporaneously admitted to hospital with similar age and vascular risk-factor profiles without AF (noAF, n = 30). Results Patients with AF have similar total GPVI surface expression ( p = 0.58) and P-selectin exposure ( p = 0.73) on their platelets compared with noAF patients but demonstrate significantly higher GPVI-dimer expression ( p = 0.02 ). Patients with paroxysmal AF express similar GPVI-dimer levels compared with permanent AF and GPVI-dimer levels were not different between anticoagulated groups. Serum N-terminal pro b-type natriuretic peptide ( p < 0.0001 ) and high sensitivity C-reactive protein ( p < 0.0001 ) were significantly correlated with GPVI-dimer expression in AF platelets. AF was the only vascular risk factor that was independently associated with higher GPVI-dimer expression in the whole population ( p = 0.02 ) . Conclusion GPVI inhibition is being explored in clinical trials as a novel target for IS treatment. As GPVI-dimer is elevated in AF patients' platelets, the exploration of targeted GPVI-dimer inhibition for stroke prevention in patients at high risk of IS due to AF is supported.

Fully automated platelet differential interference contrast image analysis via deep learning.

Platelets mediate arterial thrombosis, a leading cause of myocardial infarction and stroke. During injury, platelets adhere and spread over exposed subendothelial matrix substrates of the damaged blood vessel wall. The mechanisms which govern platelet activation and their interaction with a range of substrates are therefore regularly investigated using platelet spreading assays. These assays often use differential interference contrast (DIC) microscopy to assess platelet morphology and analysis performed using manual annotation. Here, a convolutional neural network (CNN) allowed fully automated analysis of platelet spreading assays captured by DIC microscopy. The CNN was trained using 120 generalised training images. Increasing the number of training images increases the mean average precision of the CNN. The CNN performance was compared to six manual annotators. Significant variation was observed between annotators, highlighting bias when manual analysis is performed. The CNN effectively analysed platelet morphology when platelets spread over a range of substrates (CRP-XL, vWF and fibrinogen), in the presence and absence of inhibitors (dasatinib, ibrutinib and PRT-060318) and agonist (thrombin), with results consistent in quantifying spread platelet area which is comparable to published literature. The application of a CNN enables, for the first time, automated analysis of platelet spreading assays captured by DIC microscopy.

Platelet surface receptor glycoprotein VI-dimer is overexpressed in stroke: The Glycoprotein VI in Stroke (GYPSIE) study results.

OBJECTIVES: Platelet activation underpins thrombus formation in ischemic stroke. The active, dimeric form of platelet receptor glycoprotein (GP) VI plays key roles by binding platelet ligands collagen and fibrin, leading to platelet activation. We investigated whether patients presenting with stroke expressed more GPVI on their platelet surface and had more active circulating platelets as measured by platelet P-selectin exposure. METHODS: 129 ischemic or hemorrhagic stroke patients were recruited within 8h of symptom onset. Whole blood was analyzed for platelet-surface expression of total GPVI, GPVI-dimer, and P-selectin by flow cytometry at admission and day-90 post-stroke. Results were compared against a healthy control population (n = 301). RESULTS: The platelets of stroke patients expressed significantly higher total GPVI and GPVI-dimer (P<0.0001) as well as demonstrating higher resting P-selectin exposure (P<0.0001), a measure of platelet activity, compared to the control group, suggesting increased circulating platelet activation. GPVI-dimer expression was strongly correlated circulating platelet activation [r2 = 0.88, P<0.0001] in stroke patients. Furthermore, higher platelet surface GPVI expression was associated with increased stroke severity at admission. At day-90 post-stroke, GPVI-dimer expression and was further raised compared to the level at admission (P<0.0001) despite anti-thrombotic therapy. All ischemic stroke subtypes and hemorrhagic strokes expressed significantly higher GPVI-dimer compared to controls (P<0.0001). CONCLUSIONS: Stroke patients express more GPVI-dimer on their platelet surface at presentation, lasting at least until day-90 post-stroke. Small molecule GPVI-dimer inhibitors are currently in development and the results of this study validate that GPVI-dimer as an anti-thrombotic target in ischemic stroke.

G protein-coupled receptor kinase 5 regulates thrombin signaling in platelets via PAR-1.

The interindividual variation in the functional response of platelets to activation by agonists is heritable. Genome-wide association studies (GWASs) of quantitative measures of platelet function have identified fewer than 20 distinctly associated variants, some with unknown mechanisms. Here, we report GWASs of pathway-specific functional responses to agonism by adenosine 5'-diphosphate, a glycoprotein VI-specific collagen mimetic, and thrombin receptor-agonist peptides, each specific to 1 of the G protein-coupled receptors PAR-1 and PAR-4, in subsets of 1562 individuals. We identified an association (P = 2.75 × 10-40) between a common intronic variant, rs10886430, in the G protein-coupled receptor kinase 5 gene (GRK5) and the sensitivity of platelets to activate through PAR-1. The variant resides in a megakaryocyte-specific enhancer that is bound by the transcription factors GATA1 and MEIS1. The minor allele (G) is associated with fewer GRK5 transcripts in platelets and the greater sensitivity of platelets to activate through PAR-1. We show that thrombin-mediated activation of human platelets causes binding of GRK5 to PAR-1 and that deletion of the mouse homolog Grk5 enhances thrombin-induced platelet activation sensitivity and increases platelet accumulation at the site of vascular injury. This corroborates evidence that the human G allele of rs10886430 is associated with a greater risk for cardiovascular disease. In summary, by combining the results of pathway-specific GWASs and expression quantitative trait locus studies in humans with the results from platelet function studies in Grk5-/- mice, we obtain evidence that GRK5 regulates the human platelet response to thrombin via the PAR-1 pathway.

Transcriptional, epigenetic and metabolic signatures in cardiometabolic syndrome defined by extreme phenotypes.

BACKGROUND: This work is aimed at improving the understanding of cardiometabolic syndrome pathophysiology and its relationship with thrombosis by generating a multi-omic disease signature. METHODS/RESULTS: We combined classic plasma biochemistry and plasma biomarkers with the transcriptional and epigenetic characterisation of cell types involved in thrombosis, obtained from two extreme phenotype groups (morbidly obese and lipodystrophy) and lean individuals to identify the molecular mechanisms at play, highlighting patterns of abnormal activation in innate immune phagocytic cells. Our analyses showed that extreme phenotype groups could be distinguished from lean individuals, and from each other, across all data layers. The characterisation of the same obese group, 6 months after bariatric surgery, revealed the loss of the abnormal activation of innate immune cells previously observed. However, rather than reverting to the gene expression landscape of lean individuals, this occurred via the establishment of novel gene expression landscapes. NETosis and its control mechanisms emerge amongst the pathways that show an improvement after surgical intervention. CONCLUSIONS: We showed that the morbidly obese and lipodystrophy groups, despite some differences, shared a common cardiometabolic syndrome signature. We also showed that this could be used to discriminate, amongst the normal population, those individuals with a higher likelihood of presenting with the disease, even when not displaying the classic features.

Multiparameter phenotyping of platelet reactivity for stratification of human cohorts.

Accurate and comprehensive assessment of platelet function across cohorts of donors may be key to understanding the risk of thrombotic events associated with cardiovascular disease, and, hence, to help personalize the application of antiplatelet drugs. However, platelet function tests can be difficult to perform and analyze; they also can be unreliable or uninformative and poorly standardized across studies. The Platelet Phenomic Analysis (PPAnalysis) assay and associated open-source software platform were developed in response to these challenges. PPAnalysis utilizes preprepared freeze-dried microtiter plates to provide a detailed characterization of platelet function. The automated analysis of the high-dimensional data enables the identification of subpopulations of donors with distinct platelet function phenotypes. Using this approach, we identified that the sensitivity of a donor's platelets to an agonist and their capacity to generate a functional response are distinct independent metrics of platelet reactivity. Hierarchical clustering of these metrics identified 6 subgroups with distinct platelet phenotypes within healthy cohorts, indicating that platelet reactivity does not fit into the traditional simple categories of "high" and "low" responders. These platelet phenotypes were found to exist in 2 independent cohorts of healthy donors and were stable on recall. PPAnalysis is a powerful tool for stratification of cohorts on the basis of platelet reactivity that will enable investigation of the causes and consequences of differences in platelet function and drive progress toward precision medicine.

Transcriptional characterization of human megakaryocyte polyploidization and lineage commitment.

BACKGROUND: Megakaryocytes (MKs) originate from cells immuno-phenotypically indistinguishable from hematopoietic stem cells (HSCs), bypassing intermediate progenitors. They mature within the adult bone marrow and release platelets into the circulation. Until now, there have been no transcriptional studies of primary human bone marrow MKs. OBJECTIVES: To characterize MKs and HSCs from human bone marrow using single-cell RNA sequencing, to investigate MK lineage commitment, maturation steps, and thrombopoiesis. RESULTS: We show that MKs at different levels of polyploidization exhibit distinct transcriptional states. Although high levels of platelet-specific gene expression occur in the lower ploidy classes, as polyploidization increases, gene expression is redirected toward translation and posttranslational processing transcriptional programs, in preparation for thrombopoiesis. Our findings are in keeping with studies of MK ultrastructure and supersede evidence generated using in vitro cultured MKs. Additionally, by analyzing transcriptional signatures of a single HSC, we identify two MK-biased HSC subpopulations exhibiting unique differentiation kinetics. We show that human bone marrow MKs originate from these HSC subpopulations, supporting the notion that they display priming for MK differentiation. Finally, to investigate transcriptional changes in MKs associated with stress thrombopoiesis, we analyzed bone marrow MKs from individuals with recent myocardial infarction and found a specific gene expression signature. Our data support the modulation of MK differentiation in this thrombotic state. CONCLUSIONS: Here, we use single-cell sequencing for the first time to characterize the human bone marrow MK transcriptome at different levels of polyploidization and investigate their differentiation from the HSC.

Platelet function is modified by common sequence variation in megakaryocyte super enhancers.

Linking non-coding genetic variants associated with the risk of diseases or disease-relevant traits to target genes is a crucial step to realize GWAS potential in the introduction of precision medicine. Here we set out to determine the mechanisms underpinning variant association with platelet quantitative traits using cell type-matched epigenomic data and promoter long-range interactions. We identify potential regulatory functions for 423 of 565 (75%) non-coding variants associated with platelet traits and we demonstrate, through ex vivo and proof of principle genome editing validation, that variants in super enhancers play an important role in controlling archetypical platelet functions.