Comprehensive Phenotyping Research Articles

Single-Cell Profiling Reveals Contribution of Tumor Extrinsic and Intrinsic Factors to BCMA-Targeted CAR-T Cell Efficacy in Multiple Myeloma

▪Background: BCMA CAR-T cell therapy has shown great promise in relapsed/refractory multiple myeloma (RRMM) patients, even though there is unpredictable variability in the duration and depth of response. The mechanisms behind these divergent outcomes and relapse are not well understood and heterogeneity of MM patients at the level of both tumor genomics and tumor microenvironment (TME) likely contributes to this important knowledge gap. To explore this question, we performed a longitudinal high resolution single cell genomic and proteomic analysis of bone marrow (BM) and peripheral blood (PB) samples in MM patients treated with BCMA CAR-T.Methods: Longitudinal comprehensive immune phenotyping of 3.5 million peripheral blood mononuclear cells (PBMC, CD45+CD66b-) from 11 BCMA CAR-T (idecabtagene vicleucel, ide-cel) patients was achieved via mass cytometry (CyTOF) with a panel of 39 markers. In addition, a total of 45,161 bone marrow mononuclear cells (BMMC) were analyzed from 6 patients before initiation of ide-cel therapy and at relapse by unbiased mRNA profiling via single-cell RNA-seq (scRNA-seq) using the GemCode system (10x Genomics). Downstream analysis was performed using the CATALYST and Seurat R packages, respectively. Immune cell populations are reported as % of PBMC and CD138- BMMC respectively, unless noted otherwise. Reported p values correspond to non-parametric tests or paired t test where applicable.Results: We compared baseline immune cell populations in the PB and the TME (BM) with regards to depth of CAR-T response. In PB, good responders (≥VGPR) had a higher proportion of CD8+ T cells (37% in good vs 11% in poor responders (<VGPR), p=0.08) and a lower proportion of CD14+ monocytes (30% vs 61%, p=0.28) and NK cells (2% vs 6%, p=0.08). In the TME, a similar trend was confirmed for CD8+ T cells and CD14+ monocytes. (Fig. 1A)Longitudinal analysis of PBMCs revealed phenotypic changes coinciding with CAR-T expansion; CD14+ monocytes declined from week 0 to week 4 after CAR-T infusion (40% vs 13%, p=0.04), while (non-CAR) CD8+ T cells expanded from week 0 to week 4 (32% vs 43%, p=0.15). The non-CAR CD8+ T cell expansion is characterized by differentiation towards a CD8+ effector-memory phenotype (EM, CCR7-CD45RA-) (73% vs 92% of CD8+ T cells, p=0.005). (Fig. 1B) BM samples at CAR-T relapse showed reversal of this shift: CD14+ monocyte levels remain constant or are slightly elevated, while non-CAR CD8+ T cells decrease at relapse.scRNA-seq of BMMC revealed significant gene expression changes between screening and relapse tumor samples, suggesting tumor-intrinsic factors of CAR-T response. For example, when comparing the pre and post tumor samples of a patient with durable response (PFS 652 days), we observed a significant upregulation of gene expression of pro-inflammatory chemokines (CCL3, CCL4), anti-apoptotic genes (MCL-1, FOSB, JUND), and NF-kB signaling genes (NFKBIA) in post tumor. Gene Set Enrichment Analysis (GSEA) of differentially expressed genes showed significant enrichment for TNFA signaling via NF-kB Hallmark Pathway (p.adj = 0.04). We observed similar statistically significant findings between other screening and relapse samples within our cohort, as well as upon comparison of baseline samples of poor vs good responders. (Fig. 1C, D) Thus, our data suggest that anti-apoptotic gene expression could be one of the tumor intrinsic mechanisms of CAR-T therapy resistance. Notably, we did not observe loss of BCMA expression in any tumor samples.Conclusion: Single cell immune profiling and transcriptomic sequencing highlights changes in the PB, TME and within the tumor, which in concert may influence CAR-T efficacy. Our integrated data analysis indicates general immune activation after CAR-T cell infusion that returns to baseline levels at relapse. Specifically, the expansion of non-CAR cytotoxic CD8+ EM T cells provides a rationale for co-administration of IMiDs to enhance CAR-T efficacy. Significant up-regulation of anti-apoptotic genes at baseline in poor responders, and at relapse in good responders, suggest a novel tumor-mediated escape mechanism. Targeting the MCL-1/BCL-2 axis may augment CAR-T efficacy by sensitizing tumor cells and enhancing the effect of CAR-T killing. We will confirm these findings in a longitudinal cohort of BMMC/PBMC CITE-seq patients (n=23) and will present results at the conference. [Display omitted] DisclosuresSebra: Sema4: Current Employment. Parekh: Foundation Medicine Inc: Consultancy; Amgen: Research Funding; PFIZER: Research Funding; CELGENE: Research Funding; Karyopharm Inv: Research Funding.

Assessing Cardiac Amyloidosis Subtypes by Unsupervised Phenotype Clustering Analysis

BackgroundCardiac amyloidosis (CA) is a set of amyloid diseases with usually predominant cardiac symptoms, including light-chain amyloidosis (AL), hereditary variant transthyretin amyloidosis (ATTRv), and wild-type transthyretin amyloidosis (ATTRwt). CA are characterized by high heterogeneity in phenotypes leading to diagnosis delay and worsened outcomes. ObjectivesThe authors used clustering analysis to identify typical clinical profiles in a large population of patients with suspected CA. MethodsData were collected from the French Referral Center for Cardiac Amyloidosis database (Hôpital Henri Mondor, Créteil), including 1,394 patients with suspected CA between 2010 and 2018: 345 (25%) had a diagnosis of AL, 263 (19%) ATTRv, 402 (29%) ATTRwt, and 384 (28%) no amyloidosis. Based on comprehensive clinicobiological phenotyping, unsupervised clustering analyses were performed by artificial neural network–based self-organizing maps to identify patient profiles (clusters) with similar characteristics, independent of the final diagnosis and prognosis. ResultsMean age and left ventricular ejection fraction were 72 ± 13 years and 52% ± 13%, respectively. The authors identified 7 clusters of patients with contrasting profiles and prognosis. AL patients were distinctively located within a typical cluster; ATTRv patients were distributed across 4 clusters with varying clinical presentations, 1 of which overlapped with patients without amyloidosis; interestingly, ATTRwt patients spread across 3 distinct clusters with contrasting risk factors, biological profiles, and prognosis. ConclusionsClustering analysis identified 7 clinical profiles with varying characteristics, prognosis, and associations with diagnosis. Especially in patients with ATTRwt, these results suggest key areas to improve amyloidosis diagnosis and stratify prognosis depending on associated risk factors.

Editorial introductions

Editorial introductions

Cohort profile: The Western Australian Sleep health study, a prospective sleep clinic cohort study

Genetic and epidemiologic investigations into obstructive sleep apnoea (OSA) have been limited by a scarcity of sizeable well-characterised sleep clinic cohorts with laboratory-based polysomnography (PSG). This profile reports the characteristics of a prospective clinic cohort study exploring the genotypic and phenotypic features of OSA with ongoing patient follow-up to assess long-term health outcomes. The Western Australian Sleep Health Study (WASHS) recruited patients at a large tertiary hospital sleep clinic in Perth, Australia. Between 2006 and 2010, 5948 consecutive new adult patients attended the clinic and 4914 were eligible to participate following consent and screening. Among eligible patients, 98.5% (n = 4839) had diagnostic PSG available, and 86.0% (n = 4226) were comprehensively phenotyped by clinical questionnaire and anthropometric measurements. Among those comprehensively phenotyped, blood biospecimens for biochemistry and DNA were obtained in 2759 (65.2%), and linked health administrative data was requested in 2017 for 4067 patients (96.2%). The group of most interest, the comprehensively phenotyped patients (n = 4226), were predominantly male (60.6%), middle-aged (mean±SD: 50.5 ± 14.0 years), and obese (32.7 ± 7.7 kg/m2). The majority of this group were diagnosed with OSA (93.8%). The WASHS Prospective Sleep Clinic Cohort is amongst the largest OSA cohorts globally with PSG and long-term morbidity and mortality data. Comprehensive phenotype and genotype data have contributed to numerous publications on the epidemiology and genetics of OSA. Patients have been monitored by ongoing clinic review, where OSA treatment data is collected, and by follow-up studies, such as an Australian National Health and Medical Research Council funded project (2018–2021) investigating cardiovascular outcomes in OSA.

Genome wide association mapping for agronomic, fruit quality, and root architectural traits in tomato under organic farming conditions

BackgroundOpportunity and challenges of the agriculture scenario of the next decades will face increasing demand for secure food through approaches able to minimize the input to cultivations. Large panels of tomato varieties represent a valuable resource of traits of interest under sustainable cultivation systems and for genome-wide association studies (GWAS). For mapping loci controlling the variation of agronomic, fruit quality, and root architecture traits, we used a heterogeneous set of 244 traditional and improved tomato accessions grown under organic field trials. Here we report comprehensive phenotyping and GWAS using over 37,300 SNPs obtained through double digest restriction-site associated DNA (dd-RADseq).ResultsA wide range of phenotypic diversity was observed in the studied collection, with highly significant differences encountered for most traits. A variable level of heritability was observed with values up to 69% for morphological traits while, among agronomic ones, fruit weight showed values above 80%. Genotype by environment analysis highlighted the strongest genotypic effect for aboveground traits compared to root architecture, suggesting that the hypogeal part of tomato plants has been a minor objective for breeding activities. GWAS was performed by a compressed mixed linear model leading to 59 significantly associated loci, allowing the identification of novel genes related to flower and fruit characteristics. Most genomic associations fell into the region surrounding SUN, OVATE, and MYB gene families. Six flower and fruit traits were associated with a single member of the SUN family (SLSUN31) on chromosome 11, in a region involved in the increase of fruit weight, locules number, and fruit fasciation. Furthermore, additional candidate genes for soluble solids content, fruit colour and shape were found near previously reported chromosomal regions, indicating the presence of synergic and multiple linked genes underlying the variation of these traits.ConclusionsResults of this study give new hints on the genetic basis of traits in underexplored germplasm grown under organic conditions, providing a framework for the development of markers linked to candidate genes of interest to be used in genomics-assisted breeding in tomato, in particular under low-input and organic cultivation conditions.

Comprehensive phenotyping of human peripheral blood B lymphocytes in healthy conditions

The B cell compartment provides innate and adaptive immune defenses against pathogens. Different B cell subsets, reflecting the maturation stages of B cells, have noninterchangeable functions and roles in innate and adaptive immune responses. In this review, we provide an overview of the B cell subsets present in peripheral blood of healthy individuals. A specific gating strategy is also described to clearly and univocally identify B cell subsets based on the their phenotypic traits by flow cytometric analysis.

Comprehensive Comparative Genomics and Phenotyping of Methylobacterium Species.

The pink-pigmented facultative methylotrophs (PPFMs), a major bacterial group found in the plant phyllosphere, comprise two genera: Methylobacterium and Methylorubrum. They have been separated into three major clades: A, B (Methylorubrum), and C. Within these genera, however, some species lack either pigmentation or methylotrophy, which raises the question of what actually defines the PPFMs. The present study employed a comprehensive comparative genomics approach to reveal the phylogenetic relationship among the PPFMs and to explain the genotypic differences that confer their different phenotypes. We newly sequenced the genomes of 29 relevant-type strains to complete a dataset for almost all validly published species in the genera. Through comparative analysis, we revealed that methylotrophy, nitrate utilization, and anoxygenic photosynthesis are hallmarks differentiating the PPFMs from the other Methylobacteriaceae. The Methylobacterium species in clade A, including the type species Methylobacterium organophilum, were phylogenetically classified into six subclades, each possessing relatively high genomic homology and shared phenotypic characteristics. One of these subclades is phylogenetically close to Methylorubrum species; this finding led us to reunite the two genera into a single genus Methylobacterium. Clade C, meanwhile, is composed of phylogenetically distinct species that share relatively higher percent G+C content and larger genome sizes, including larger numbers of secondary metabolite clusters. Most species of clade C and some of clade A have the glutathione-dependent pathway for formaldehyde oxidation in addition to the H4MPT pathway. Some species cannot utilize methanol due to their lack of MxaF-type methanol dehydrogenase (MDH), but most harbor an XoxF-type MDH that enables growth on methanol in the presence of lanthanum. The genomes of PPFMs encode between two and seven (average 3.7) genes for pyrroloquinoline quinone-dependent alcohol dehydrogenases, and their phylogeny is distinctly correlated with their genomic phylogeny. All PPFMs were capable of synthesizing auxin and did not induce any immune response in rice cells. Other phenotypes including sugar utilization, antibiotic resistance, and antifungal activity correlated with their phylogenetic relationship. This study provides the first inclusive genotypic insight into the phylogeny and phenotypes of PPFMs.

Neuronal cholesterol synthesis is essential for repair of chronically demyelinated lesions in mice.

Astrocyte-derived cholesterol supports brain cells under physiological conditions. However, in demyelinating lesions, astrocytes downregulate cholesterol synthesis, and the cholesterol that is essential for remyelination has to originate from other cellular sources. Here, we show that repair following acute versus chronic demyelination involves distinct processes. In particular, in chronic myelin disease, when recycling of lipids is often defective, de novo neuronal cholesterol synthesis is critical for regeneration. By gene expression profiling, genetic loss-of-function experiments, and comprehensive phenotyping, we provide evidence that neurons increase cholesterol synthesis in chronic myelin disease models and in patients with multiple sclerosis (MS). In mouse models, neuronal cholesterol facilitates remyelination specifically by triggering oligodendrocyte precursor cell proliferation. Our data contribute to the understanding of disease progression and have implications for therapeutic strategies in patients with MS.

A comprehensive digital phenotype for postpartum hemorrhage.

ObjectiveWe aimed to establish a comprehensive digital phenotype for postpartum hemorrhage (PPH). Current guidelines rely primarily on estimates of blood loss, which can be inaccurate and biased and ignore complementary information readily available in electronic medical records (EMR). Inaccurate and incomplete phenotyping contributes to ongoing challenges in tracking PPH outcomes, developing more accurate risk assessments, and identifying novel interventions.Materials and MethodsWe constructed a cohort of 71 944 deliveries from the Mount Sinai Health System. Estimates of postpartum blood loss, shifts in hematocrit, administration of uterotonics, surgical interventions, and diagnostic codes were combined to identify PPH, retrospectively. Clinical features were extracted from EMRs and mapped to common data models for maximum interoperability across hospitals. Blinded chart review was done by a physician on a subset of PPH and non-PPH patients and performance was compared to alternate PPH phenotypes. PPH was defined as clinical diagnosis of postpartum hemorrhage documented in the patient’s chart upon chart review.ResultsWe identified 6639 PPH deliveries (9% prevalence) using our phenotype—more than 3 times as many as using blood loss alone (N = 1,747), supporting the need to incorporate other diagnostic and intervention data. Chart review revealed our phenotype had 89% accuracy and an F1-score of 0.92. Alternate phenotypes were less accurate, including a common blood loss-based definition (67%) and a previously published digital phenotype (74%).ConclusionWe have developed a scalable, accurate, and valid digital phenotype that may be of significant use for tracking outcomes and ongoing clinical research to deliver better preventative interventions for PPH.

Upper gastrointestinal tract condition in patients with morbid obesity before and after bariatric surgery

Obesity is the 21st century pandemic. By 2025 6 % of men and 9 % of women will suffer from morbid obesity. For morbid obesity bariatric surgery is the main treatment option. The presence of pathology of the upper gastrointestinal tract (GERD) determines the choice of surgical intervention. The study involved 68 patients 47.1 ± 10.2 years with morbid obesity. Patients had the following gastroenterological pathologies: erosive esophagitis – 6 (8.8 %), hiatal hernia – 9 (13 %), erosive and ulcerative lesions of the stomach and duodenum – 13 (19 %). Heartburn disturbed 38 (55 %), regurgitation – 17 (25 %), severity and feeling of fullness in the epigastrium – 29 (42 %), pain or burning in the epigastrium – 8 (11 %) patients. Fifty (73.5 %) patients underwent laparoscopic sleeve gastrectomy, 18 (26.5 %) patients – laparoscopic Roux-en-Y gastric bypass. Comprehensive preoperative examination (x-ray of the upper gastrointestinal tract with barium, daily pH impedance measurement, esophageal manometry) and phenotyping of GERD lead to reduction of GERD symptoms in patients with morbid obesity after sleeve gastrectomy.

Comprehensive phenotyping of peripheral blood monocytes in healthy bovine.

Monocytes are bone marrow derived innate myeloid cells that circulate in the blood and play important roles in infection and inflammation. As part of the mononuclear phagocytic system, monocytes provide innate effector functions, support the adaptive immune response, and play a role in the maintenance of tissue homeostasis. In addition to their role in sensing pathogen-associated molecular patterns using several pattern recognition receptors, monocytes are characterized by their ability to ingest and kill microbes, to produce cytokines and chemokines, and to present antigens to T cells. For a long time, monocytes have been considered as a homogenous cell population, characterized by the expression of CD14, the receptor of lipopolysaccharide. Studies in several species have shown that the monocyte population consists of phenotypically and functionally different cell subsets. In this review, we report a comprehensive phenotyping of monocyte subsets in cattle. In addition, the most characterizing cell markers and gating strategies for detailed immunophenotyping of bovine monocyte subsets are discussed.

The Monash Autism-ADHD genetics and neurodevelopment (MAGNET) project design and methodologies: a dimensional approach to understanding neurobiological and genetic aetiology

BackgroundASD and ADHD are prevalent neurodevelopmental disorders that frequently co-occur and have strong evidence for a degree of shared genetic aetiology. Behavioural and neurocognitive heterogeneity in ASD and ADHD has hampered attempts to map the underlying genetics and neurobiology, predict intervention response, and improve diagnostic accuracy. Moving away from categorical conceptualisations of psychopathology to a dimensional approach is anticipated to facilitate discovery of data-driven clusters and enhance our understanding of the neurobiological and genetic aetiology of these conditions. The Monash Autism-ADHD genetics and neurodevelopment (MAGNET) project is one of the first large-scale, family-based studies to take a truly transdiagnostic approach to ASD and ADHD. Using a comprehensive phenotyping protocol capturing dimensional traits central to ASD and ADHD, the MAGNET project aims to identify data-driven clusters across ADHD-ASD spectra using deep phenotyping of symptoms and behaviours; investigate the degree of familiality for different dimensional ASD-ADHD phenotypes and clusters; and map the neurocognitive, brain imaging, and genetic correlates of these data-driven symptom-based clusters.MethodsThe MAGNET project will recruit 1,200 families with children who are either typically developing, or who display elevated ASD, ADHD, or ASD-ADHD traits, in addition to affected and unaffected biological siblings of probands, and parents. All children will be comprehensively phenotyped for behavioural symptoms, comorbidities, neurocognitive and neuroimaging traits and genetics.ConclusionThe MAGNET project will be the first large-scale family study to take a transdiagnostic approach to ASD-ADHD, utilising deep phenotyping across behavioural, neurocognitive, brain imaging and genetic measures.

Intravascular molecular‐structural imaging with a miniaturized integrated near‐infrared fluorescence and ultrasound catheter

Coronary artery disease (CAD) remains a leading cause of mortality and warrants new imaging approaches to better guide clinical care. We report on a miniaturized, hybrid intravascular catheter and imaging system for comprehensive coronary artery imaging in vivo. Our catheter exhibits a total diameter of 1.0 mm (3.0 French), equivalent to standalone clinical intravascular ultrasound (IVUS) catheters but enables simultaneous near-infrared fluorescence (NIRF) and IVUS molecular-structural imaging. We demonstrate NIRF-IVUS imaging in vitro in coronary stents using NIR fluorophores, and compare NIRF signal strengths for prism and ball lens sensor designs in both low and high scattering media. Next, in vivo intravascular imaging in pig coronary arteries demonstrates simultaneous, co-registered molecular-structural imaging of experimental CAD inflammation on IVUS and distance-corrected NIRF images. The obtained results suggest substantial potential for the NIRF-IVUS catheter to advance standalone IVUS, and enable comprehensive phenotyping of vascular disease to better assess and treat patients with CAD.

Genome-Wide Association Study Identifies Risk Loci for Cluster Headache.

This study was undertaken to identify susceptibility loci for cluster headache and obtain insights into relevant disease pathways. We carried out a genome-wide association study, where 852 UK and 591 Swedish cluster headache cases were compared with 5,614 and 1,134 controls, respectively. Following quality control and imputation, single variant association testing was conducted using a logistic mixed model for each cohort. The 2 cohorts were subsequently combined in a merged analysis. Downstream analyses, such as gene-set enrichment, functional variant annotation, prediction and pathway analyses, were performed. Initial independent analysis identified 2 replicable cluster headache susceptibility loci on chromosome 2. A merged analysis identified an additional locus on chromosome 1 and confirmed a locus significant in the UK analysis on chromosome 6, which overlaps with a previously known migraine locus. The lead single nucleotide polymorphisms were rs113658130 (p=1.92× 10-17 , odds ratio [OR]=1.51, 95% confidence interval [CI] = 1.37-1.66) and rs4519530 (p=6.98× 10-17 , OR=1.47, 95% CI=1.34-1.61) on chromosome 2, rs12121134 on chromosome 1 (p=1.66× 10-8 , OR=1.36, 95% CI=1.22-1.52), and rs11153082 (p=1.85× 10-8 , OR=1.30, 95% CI=1.19-1.42) on chromosome 6. Downstream analyses implicated immunological processes in the pathogenesis of cluster headache. We identified and replicated several genome-wide significant associations supporting a genetic predisposition in cluster headache in a genome-wide association study involving 1,443 cases. Replication in larger independent cohorts combined with comprehensive phenotyping, in relation to, for example, treatment response and cluster headache subtypes, could provide unprecedented insights into genotype-phenotype correlations and the pathophysiological pathways underlying cluster headache. ANN NEUROL 2021;90:193-202.

Tissue-specific expression of Cas9 has no impact on whole-body metabolism in four transgenic mouse lines

ObjectiveCRISPR/Cas9 technology has revolutionized gene editing and fast tracked our capacity to manipulate genes of interest for the benefit of both research and therapeutic applications. Whilst many advances have, and continue to be made in this area, perhaps the most utilized technology to date has been the generation of knockout cells, tissues and animals. The advantages of this technology are many fold, however some questions still remain regarding the effects that long term expression of foreign proteins such as Cas9, have on mammalian cell function. Several studies have proposed that chronic overexpression of Cas9, with or without its accompanying guide RNAs, may have deleterious effects on cell function and health. This is of particular concern when applying this technology in vivo, where chronic expression of Cas9 in tissues of interest may promote disease-like phenotypes and thus confound the investigation of the effects of the gene of interest. Although these concerns remain valid, no study to our knowledge has yet to demonstrate this directly. MethodsIn this study we used the lox-stop-lox (LSL) spCas9 ROSA26 transgenic (Tg) mouse line to generate four tissue-specific Cas9-Tg models that express Cas9 in the heart, liver, skeletal muscle or adipose tissue. We performed comprehensive phenotyping of these mice up to 20-weeks of age and subsequently performed molecular analysis of their organs. ResultsWe demonstrate that Cas9 expression in these tissues had no detrimental effect on whole body health of the animals, nor did it induce any tissue-specific effects on whole body energy metabolism, liver health, inflammation, fibrosis, heart function or muscle mass. ConclusionsOur data suggests that these models are suitable for studying the tissue specific effects of gene deletion using the LSL-Cas9-Tg model, and that phenotypes observed utilizing these models can be confidently interpreted as being gene specific, and not confounded by the chronic overexpression of Cas9.

AI-enabled Comprehensive Coronary Phenotyping In Patients Being Referred For Invasive Coronary Angiography After Abnormal Stress Testing: A Randomized Controlled Trial

Introduction: In patients with suspected stable coronary artery disease, non-invasive ischemia test (NIST) use frequently has been associated with high normalcy rates for individuals referred for invasive coronary angiography (ICA). This study aimed to determine the impact of an artificial intelligence (AI)-enabled computed tomography angiography solution for comprehensive coronary artery evaluation on patient selection for ICA after abnormal NIST.

How Somatic Mutations Carve Morphologic Imprint in MDS?

Next generation sequencing (NGS) investigations have identified individual mutations that have been useful in differential diagnoses and as biomarkers of prognosis. These tasks are complicated by the heterogeneity of mutations, the overlap of their spectra between disease subtypes, their combinations, and their clonal architecture. All of this drives diversity and obscures individual genotype/phenotype associations. Our goal was to develop robust new approaches to assessing such phenotype/genotype associations.Our focus is on the relationship between cell morphology and counts and mutations and their combinations. We hypothesized that while MDS morphology is heterogeneous, it must be result molecular lesions occurring in particular hierarchical relationships. Based on this we examined 681 cases of MDS that were comprehensively annotated in terms of the presence of diagnostic morphologic abnormalities; these were assessed by an unbiased pathologist who uniformly used standard diagnostic criteria. Abnormalities included the presence of fibrosis, individual lineage dysplasia, cytopenia, and increases in monocyte and megakaryocyte counts. Correlations between these disease features and mutational profiles, considering clonal dominant and secondary events separately, were then assessed. The bio-analytic platform used generated 3-dimensional plots of odds ratios of changes in morphologies, cell counts, and stage, which were estimated using logistic regression cubes, in which the odds of association with particular morphological changes have been plotted for each mutation. Similar 3D regression cubes have been generated for cytopenias or stage of the disease. This analysis included initial hits and animation to visualize the impact of subsequent hits on lineage dysplasia, myeloproliferation, and % blasts.Along the continuum between myelodysplasia/myeloproliferation and acute leukemia, 32% (219/681) of patients had trilineage dysplasia. Myeloid, erythroid and megakaryocyte dysplasia occurred in 52% (356/681), 67% (453/681) and, 66% (449/681) of patients, respectively. Monocytosis (>1000/uL) was present in 20% of the patients and 16% of cases displayed a combination of at least 2 lineage dysplasia and myeloproliferative features. Using targeted NGS on a panel of genes often mutated in MDS, ancestral and secondary hits were identified and their impact on phenotype was assessed. For example, ASXL1, SF3B1, and SRSF2 resultinuni-lineage dysplasia (n=155) compared to unilineage MDS cases without dysplasia, and ASXL1, SRSF2, and TP53 mutations are significantly affected in multi-lineage dysplasia (n=219), each having a different, defined association with myeloproliferative features.Simultaneous analysis of these influences resulted in a number of salient relationships that illustrate how specific phenotypic features evolve in a deterministic molecular context: TET2 mutations confer myelodysplastic and monocytic character; SRSF2 mutations confer myeloid and myeloproliferative character that includes megakaryocytic hyper/dysplasia; mutations in ASXL1, RUNX1, and TP53 impart myelodysplasia; SF3B1 and DNMT3A mutations effected erythroid morphology (Figure 1); SRSF2 and ASXL1 mutations affect myeloid and megakaryocyte lineage; and TP53 mutated cases showed significantly higher frequency in multi-lineage dysplasia. Focusing on cytopenia, TP53 mutations significantly correlated with leukopenia and thrombocytopenia, U2AF1 mutations associate with anemia, RUNX1 and ASXL1 mutations associate with thrombocytopenia, and mutations in TET2, ASXL1, SRSF2 and RUNX1 associate with increased odds of myeloproliferation, particularly monocytosis . We then applied this approach to ancestral events. In 3D, we were able to visualize how common secondary hits shape the phenotype landscape of progression and concomitant cytopenia.Our report results of a comprehensive genotype phenotype association study that linked somatic molecular events to clinical features. We demonstrated that with tools and sufficient statistical power, complexities of how molecular context governs morphologic changes can be deciphered. [Display omitted] DisclosuresSekeres:Celgene: Membership on an entity's Board of Directors or advisory committees.

OmiEmbed: A Unified Multi-Task Deep Learning Framework for Multi-Omics Data

Simple SummaryOmiEmbed is a unified multi-task deep learning framework for multi-omics data, supporting dimensionality reduction, multi-omics integration, tumour type classification, phenotypic feature reconstruction and survival prediction. The framework is comprised of deep embedding and downstream task modules to capture biomedical information from high-dimensional omics data. OmiEmbed outperformed state-of-the-art methods on all three types of downstream tasks: classification, regression and survival prediction. Better performance was achieved using the multi-task training strategy compared to training each downstream task individually. With the multi-task strategy, OmiEmbed learnt a comprehensive omics embedding containing information from multiple tasks. OmiEmbed is open source, well-organised and convenient to be extended to other customised input data, network structures and downstream tasks, which has promising potential to facilitate more accurate and personalised clinical decision making.High-dimensional omics data contain intrinsic biomedical information that is crucial for personalised medicine. Nevertheless, it is challenging to capture them from the genome-wide data, due to the large number of molecular features and small number of available samples, which is also called “the curse of dimensionality” in machine learning. To tackle this problem and pave the way for machine learning-aided precision medicine, we proposed a unified multi-task deep learning framework named OmiEmbed to capture biomedical information from high-dimensional omics data with the deep embedding and downstream task modules. The deep embedding module learnt an omics embedding that mapped multiple omics data types into a latent space with lower dimensionality. Based on the new representation of multi-omics data, different downstream task modules were trained simultaneously and efficiently with the multi-task strategy to predict the comprehensive phenotype profile of each sample. OmiEmbed supports multiple tasks for omics data including dimensionality reduction, tumour type classification, multi-omics integration, demographic and clinical feature reconstruction, and survival prediction. The framework outperformed other methods on all three types of downstream tasks and achieved better performance with the multi-task strategy compared to training them individually. OmiEmbed is a powerful and unified framework that can be widely adapted to various applications of high-dimensional omics data and has great potential to facilitate more accurate and personalised clinical decision making.

Comprehensive echocardiographic evaluation of the right heart in patients with pulmonary vascular diseases: the PVDOMICS experience.

There is a wide spectrum of diseases associated with pulmonary hypertension, pulmonary vascular remodelling, and right ventricular dysfunction. The NIH-sponsored PVDOMICS network seeks to perform comprehensive clinical phenotyping and endophenotyping across these disorders to further evaluate and define pulmonary vascular disease. Echocardiography represents the primary non-invasive method to phenotype cardiac anatomy, function, and haemodynamics in these complex patients. However, comprehensive right heart evaluation requires the use of multiple echocardiographic parameters and optimized techniques to ensure optimal image acquisition. The PVDOMICS echo protocol outlines the best practice approach to echo phenotypic assessment of the right heart/pulmonary artery unit. Novel workflow processes, methods for quality control, data for feasibility of measurements, and reproducibility of right heart parameters derived from this study provide a benchmark frame of reference. Lessons learned from this protocol will serve as a best practice guide for echocardiographic image acquisition and analysis across the spectrum of right heart/pulmonary vascular disease.

The hepcidin regulator erythroferrone is a new member of the erythropoiesis-iron-bone circuitry.

Erythroblast erythroferrone (ERFE) secretion inhibits hepcidin expression by sequestering several bone morphogenetic protein (BMP) family members to increase iron availability for erythropoiesis. To address whether ERFE functions also in bone and whether the mechanism of ERFE action in bone involves BMPs, we utilize the Erfe-/- mouse model as well as β-thalassemic (Hbbth3/+) mice with systemic loss of ERFE expression. In additional, we employ comprehensive skeletal phenotyping analyses as well as functional assays in vitro to address mechanistically the function of ERFE in bone. We report that ERFE expression in osteoblasts is higher compared with erythroblasts, is independent of erythropoietin, and functional in suppressing hepatocyte hepcidin expression. Erfe-/- mice display low-bone-mass arising from increased bone resorption despite a concomitant increase in bone formation. Consistently, Erfe-/- osteoblasts exhibit enhanced mineralization, Sost and Rankl expression, and BMP-mediated signaling ex vivo. The ERFE effect on osteoclasts is mediated through increased osteoblastic RANKL and sclerostin expression, increasing osteoclastogenesis in Erfe-/- mice. Importantly, Erfe loss in Hbbth3/+mice, a disease model with increased ERFE expression, triggers profound osteoclastic bone resorption and bone loss. Together, ERFE exerts an osteoprotective effect by modulating BMP signaling in osteoblasts, decreasing RANKL production to limit osteoclastogenesis, and prevents excessive bone loss during expanded erythropoiesis in β-thalassemia. YZG acknowledges the support of the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) (R01 DK107670 to YZG and DK095112 to RF, SR, and YZG). MZ acknowledges the support of the National Institute on Aging (U19 AG60917) and NIDDK (R01 DK113627). TY acknowledges the support of the National Institute on Aging (R01 AG71870). SR acknowledges the support of NIDDK (R01 DK090554) and Commonwealth Universal Research Enhancement (CURE) Program Pennsylvania.