Validating and automating learning of cardiometabolic polygenic risk scores from direct-to-consumer genetic and phenotypic data: implications for scaling precision health research.

INTRODUCTION: A major challenge to enabling precision health at a global scale is the bias between those who enroll in state sponsored genomic research and those suffering from chronic disease. More than 30 million people have been genotyped by direct-to-consumer (DTC) companies such as 23andMe, Ancestry DNA, and MyHeritage, providing a potential mechanism for democratizing access to medical interventions and thus catalyzing improvements in patient outcomes as the cost of data acquisition drops. However, much of these data are sequestered in the initial provider network, without the ability for the scientific community to either access or validate. Here, we present a novel geno-pheno platform that integrates heterogeneous data sources and applies learnings to common chronic disease conditions including Type 2 diabetes (T2D) and hypertension. METHODS: We collected genotyped data from a novel DTC platform where participants upload their genotype data files and were invited to answer general health questionnaires regarding cardiometabolic traits over a period of 6 months. Quality control, imputation, and genome-wide association studies were performed on this dataset, and polygenic risk scores were built in a case-control setting using the BASIL algorithm. RESULTS: We collected data on N = 4,550 (389 cases / 4,161 controls) who reported being affected or previously affected for T2D and N = 4,528 (1,027 cases / 3,501 controls) for hypertension. We identified 164 out of 272 variants showing identical effect direction to previously reported genome-significant findings in Europeans. Performance metric of the PRS models was AUC = 0.68, which is comparable to previously published PRS models obtained with larger datasets including clinical biomarkers. DISCUSSION: DTC platforms have the potential of inverting research models of genome sequencing and phenotypic data acquisition. Quality control (QC) mechanisms proved to successfully enable traditional GWAS and PRS analyses. The direct participation of individuals has shown the potential to generate rich datasets enabling the creation of PRS cardiometabolic models. More importantly, federated learning of PRS from reuse of DTC data provides a mechanism for scaling precision health care delivery beyond the small number of countries who can afford to finance these efforts directly. CONCLUSIONS: The genetics of T2D and hypertension have been studied extensively in controlled datasets, and various polygenic risk scores (PRS) have been developed. We developed predictive tools for both phenotypes trained with heterogeneous genotypic and phenotypic data generated outside of the clinical environment and show that our methods can recapitulate prior findings with fidelity. From these observations, we conclude that it is possible to leverage DTC genetic repositories to identify individuals at risk of debilitating diseases based on their unique genetic landscape so that informed, timely clinical interventions can be incorporated.

HDAC-regulated myomiRs control BAF60 variant exchange and direct the functional phenotype of fibro-adipogenic progenitors in dystrophic muscles.

Fibro-adipogenic progenitors (FAPs) are important components of the skeletal muscle regenerative environment. Whether FAPs support muscle regeneration or promote fibro-adipogenic degeneration is emerging as a key determinant in the pathogenesis of muscular diseases, including Duchenne muscular dystrophy (DMD). However, the molecular mechanism that controls FAP lineage commitment and activity is currently unknown. We show here that an HDAC-myomiR-BAF60 variant network regulates the fate of FAPs in dystrophic muscles of mdx mice. Combinatorial analysis of gene expression microarray, genome-wide chromatin remodeling by nuclease accessibility (NA) combined with next-generation sequencing (NA-seq), small RNA sequencing (RNA-seq), and microRNA (miR) high-throughput screening (HTS) against SWI/SNF BAF60 variants revealed that HDAC inhibitors (HDACis) derepress a "latent" myogenic program in FAPs from dystrophic muscles at early stages of disease. Specifically, HDAC inhibition induces two core components of the myogenic transcriptional machinery, MYOD and BAF60C, and up-regulates the myogenic miRs (myomiRs) (miR-1.2, miR-133, and miR-206), which target the alternative BAF60 variants BAF60A and BAF60B, ultimately directing promyogenic differentiation while suppressing the fibro-adipogenic phenotype. In contrast, FAPs from late stage dystrophic muscles are resistant to HDACi-induced chromatin remodeling at myogenic loci and fail to activate the promyogenic phenotype. These results reveal a previously unappreciated disease stage-specific bipotency of mesenchimal cells within the regenerative environment of dystrophic muscles. Resolution of such bipotency by epigenetic intervention with HDACis provides a molecular rationale for the in situ reprogramming of target cells to promote therapeutic regeneration of dystrophic muscles.

Inhibition of miR-25 improves cardiac contractility in the failing heart.

Heart failure is characterized by a debilitating decline in cardiac function, and recent clinical trial results indicate that improving the contractility of heart muscle cells by boosting intracellular calcium handling might be an effective therapy. MicroRNAs (miRNAs) are dysregulated in heart failure but whether they control contractility or constitute therapeutic targets remains speculative. Using high-throughput functional screening of the human microRNAome, here we identify miRNAs that suppress intracellular calcium handling in heart muscle by interacting with messenger RNA encoding the sarcoplasmic reticulum calcium uptake pump SERCA2a (also known as ATP2A2). Of 875 miRNAs tested, miR-25 potently delayed calcium uptake kinetics in cardiomyocytes in vitro and was upregulated in heart failure, both in mice and humans. Whereas adeno-associated virus 9 (AAV9)-mediated overexpression of miR-25 in vivo resulted in a significant loss of contractile function, injection of an antisense oligonucleotide (antagomiR) against miR-25 markedly halted established heart failure in a mouse model, improving cardiac function and survival relative to a control antagomiR oligonucleotide. These data reveal that increased expression of endogenous miR-25 contributes to declining cardiac function during heart failure and suggest that it might be targeted therapeutically to restore function.

miR-132/212 Impairs Cardiomyocytes Contractility in the Failing Heart by Suppressing SERCA2a.

Compromised cardiac function is a hallmark for heart failure, mostly appearing as decreased contractile capacity due to dysregulated calcium handling. Unfortunately, the underlying mechanism causing impaired calcium handling is still not fully understood. Previously the miR-132/212 family was identified as a regulator of cardiac function in the failing mouse heart, and pharmaceutically inhibition of miR-132 is beneficial for heart failure. In this study, we further investigated the molecular mechanisms of miR-132/212 in modulating cardiomyocyte contractility in the context of the pathological progression of heart failure. We found that upregulated miR-132/212 expressions in all examined hypertrophic heart failure mice models. The overexpression of miR-132/212 prolongs calcium decay in isolated neonatal rat cardiomyocytes, whereas cardiomyocytes isolated from miR-132/212 KO mice display enhanced contractility in comparison to wild type controls. In response to chronic pressure-overload, miR-132/212 KO mice exhibited a blunted deterioration of cardiac function. Using a combination of biochemical approaches and in vitro assays, we confirmed that miR-132/212 regulates SERCA2a by targeting the 3'-end untranslated region of SERCA2a. Additionally, we also confirmed PTEN as a direct target of miR-132/212 and potentially participates in the cardiac response to miR132/212. In end-stage heart failure patients, miR-132/212 is upregulated and correlates with reduced SERCA2a expression. The up-regulation of miR-132/212 in heart failure impairs cardiac contractile function by targeting SERCA2a, suggesting that pharmaceutical inhibition of miR-132/212 might be a promising therapeutic approach to promote cardiac function in heart failure patients.

ErbB2 and ErbB3 regulate amputation-induced proliferation and migration during vertebrate regeneration.

Epimorphic regeneration is a unique and complex instance of postembryonic growth observed in certain metazoans that is usually triggered by severe injury [Akimenko et al., 2003; Alvarado and Tsonis, 2006; Brockes, 1997; Endo et al., 2004]. Cell division and migration are two fundamental biological processes required for supplying replacement cells during regeneration [Endo et al., 2004; Slack, 2007]. However, the connection between the early stimuli generated after injury and the signals regulating proliferation and migration during regeneration remain largely unknown. Here we show that the oncogenes ErbB2 and ErbB3, two members of the EGFR family, are essential for mounting a successful regeneration response in vertebrates. Importantly, amputation-induced progenitor proliferation and migration are significantly reduced upon genetic and/or chemical modulation of ErbB function. Moreover, we also found that NRG1 and PI3K functionally interact with ErbB2 and ErbB3 during regeneration and interfering with their function also abrogates the capacity of progenitor cells to regenerate lost structures upon amputation. Our findings suggest that ErbB, PI3K and NRG1 are components of a permissive switch for migration and proliferation continuously acting across the amputated fin from early stages of vertebrate regeneration onwards that regulate the expression of the transcription factors lef1 and msxB.

Integrating omics into the cardiac differentiation of human pluripotent stem cells.

Time-dependent extracellular manipulations of human pluripotent stem cells can yield as much as 90% pure populations of cardiomyocytes. While the extracellular control of differentiation generally entails dynamic regulation of well-known pathways such as Wnt, BMP, and Nodal signaling, the underlying genetic networks are far more complex and are poorly understood. Notably, the identification of these networks holds promise for understanding heart disease and regeneration. The availability of genome-wide experimentation, such as RNA and DNA sequencing, as well as high throughput surveying with small molecule and small interfering RNA libraries, now enables us to map the genetic interactions underlying cardiac differentiation on a global scale. Initial studies demonstrate the complexity of the genetic regulation of cardiac differentiation, exposing unanticipated novel mechanisms. However, the large datasets generated tend to be overwhelming and systematic approaches are needed to process the vast amount of data to improve our mechanistic understanding of the complex biology. Systems biology methods spur high hopes for parsing vast amounts of data into genetic interaction models that can be verified experimentally and ultimately yield functional networks that expose the genetic connections underlying biological processes.

Bioelectricity and epimorphic regeneration.

All cells have electric potentials across their membranes, but is there really compelling evidence to think that such potentials are used as instructional cues in developmental biology? Numerous reports indicate that, in fact, steady, weak bioelectric fields are observed throughout biology and function during diverse biological processes, including development. Bioelectric fields, generated upon amputation, are also likely to play a key role during vertebrate regeneration by providing the instructive cues needed to direct migrating cells to form a wound epithelium, a structure unique to regenerating animals. However, mechanistic insight is still sorely lacking in the field. What are the genes required for bioelectric-dependent cell migration during regeneration? The power of genetics combined with the use of zebrafish offers the best opportunity for unbiased identification of the molecular players in bioelectricity.

chokh/rx3 specifies the retinal pigment epithelium fate independently of eye morphogenesis.

Despite the importance of the retinal pigment epithelium (RPE) for vision, the molecular processes involved in its specification are poorly understood. We identified two new mutant alleles for the zebrafish gene chokh (chk), which display a reduction or absence of the RPE. Unexpectedly, the neural retina (NR) in chk is specified and laminated, indicating that the regulatory network leading to NR development is largely independent of the RPE. Genetic mapping and molecular characterization revealed that chk encodes Rx3. Expression analyses show that otx2 and mitfb are not expressed in the prospective RPE of chk, indicating that the retinal homeobox gene rx3 acts upstream of the molecular network controlling RPE specification. Cellular transplantations demonstrate that rx3 function is autonomously required to specify the prospective RPE. Though rx2 is also absent in chk, neither rx2 nor rx1 is required for RPE development. Thus, our data provide the first indication that, in addition to controlling optic lobe evagination and proliferation, chk/rx3 also determines cellular fate.