Abstract: N6-methyladenosine (m6A) is identified as the most common, abundant and conserved internal transcriptional modification, especially within eukaryotic messenger RNAs (mRNAs). M6A modification is installed by the m6A methyltransferases (METTL3/14, WTAP, RBM15/15B and KIAA1429, termed as “writers”), reverted by the demethylases (FTO and ALKBH5, termed as “erasers”) and recognized by m6A binding proteins (YTHDF1/2/3, IGF2BP1 and HNRNPA2B1, termed as “readers”). Acumulating evidence shows that, m6A RNA methylation has an outsize effect on RNA production/metabolism and participates in the pathogenesis of multiple diseases including cancers. Until now, the molecular mechanisms underlying m6A RNA methylation in various tumors have not been comprehensively clarified. In this review, we mainly summarize the recent advances in biological function of m6A modifications in human cancer and discuss the potential therapeutic strategies.

Abstract: A coordinated, multidisciplinary approach to care is essential for optimum management of the primary manifestations and secondary complications of Duchenne muscular dystrophy (DMD). Contemporary care has been shaped by the availability of more sensitive diagnostic techniques and the earlier use of therapeutic interventions, which have the potential to improve patients' duration and quality of life. In part 2 of this update of the DMD care considerations, we present the latest recommendations for respiratory, cardiac, bone health and osteoporosis, and orthopaedic and surgical management for boys and men with DMD. Additionally, we provide guidance on cardiac management for female carriers of a disease-causing mutation. The new care considerations acknowledge the effects of long-term glucocorticoid use on the natural history of DMD, and the need for care guidance across the lifespan as patients live longer. The management of DMD looks set to change substantially as new genetic and molecular therapies become available.

Abstract: We consider the prediction of interfaces between proteins, a challenging problem with important applications in drug discovery and design, and examine the performance of existing and newly proposed spatial graph convolution operators for this task. By performing convolution over a local neighborhood of a node of interest, we are able to stack multiple layers of convolution and learn effective latent representations that integrate information across the graph that represent the three dimensional structure of a protein of interest. An architecture that combines the learned features across pairs of proteins is then used to classify pairs of amino acid residues as part of an interface or not. In our experiments, several graph convolution operators yielded accuracy that is better than the state-of-the-art SVM method in this task.

Abstract: Recombination is a prominent feature of meiosis in which it plays an important role in increasing genetic diversity during inheritance. Additionally, in most organisms, recombination also plays mechanical roles in chromosomal processes, most notably to mediate pairing of homologous chromosomes during prophase and, ultimately, to ensure regular segregation of homologous chromosomes when they separate at the first meiotic division. Recombinational interactions are also subject to important spatial patterning at both early and late stages. Recombination-mediated processes occur in physical and functional linkage with meiotic axial chromosome structure, with interplay in both directions, before, during, and after formation and dissolution of the synaptonemal complex (SC), a highly conserved meiosis-specific structure that links homolog axes along their lengths. These diverse processes also are integrated with recombination-independent interactions between homologous chromosomes, nonhomology-based chromosome couplings/clusterings, and diverse types of chromosome movement. This review provides an overview of these diverse processes and their interrelationships.

Abstract: Background The global burden of periodontal diseases remains high. Population growth trends, changes in risk factors and improved tooth retention will increase the socio-economic burden of periodontitis that is responsible for 3,5 million years lived with disability, 54 billion USD/year in lost productivity and a major portion of the 442 billion USD/year cost for oral diseases. Methods In the context of the Milan World Exhibition 2015 “Feeding the Planet, Energy for Life”, a green paper was developed and offered for global consultation by the European Federation of Periodontology. The final draft was endorsed by professional organizations around the world and is presented to stakeholders as a call for global action. Results Specific actions for the public, policymakers, educators, and professional organizations have been identified in the areas of prevention, detection and care. These actions align public interest and knowledge, need for self-care, professional intervention and policies to the best scientific evidence to proactively promote periodontal health and effectively manage the global burden of periodontal diseases, in accordance with WHO/UN priorities and strategies for tackling common non-communicable diseases (NCDs) via the Common Risk Factor Approach. Conclusions A strong and coherent body of evidence allows identification of actionable preventive, diagnostic and therapeutic strategies to effectively promote periodontal health and general wellbeing, and better manage the socio-economic consequences. Action requires consideration of the specific national scenarios. This article is protected by copyright. All rights reserved.

Abstract: China has recently made available hourly air pollution data from over 1500 sites, including airborne particulate matter (PM), SO2, NO2, and O3. We apply Kriging interpolation to four months of data to derive pollution maps for eastern China. Consistent with prior findings, the greatest pollution occurs in the east, but significant levels are widespread across northern and central China and are not limited to major cities or geologic basins. Sources of pollution are widespread, but are particularly intense in a northeast corridor that extends from near Shanghai to north of Beijing. During our analysis period, 92% of the population of China experienced >120 hours of unhealthy air (US EPA standard), and 38% experienced average concentrations that were unhealthy. China’s population-weighted average exposure to PM2.5 was 52 μg/m3. The observed air pollution is calculated to contribute to 1.6 million deaths/year in China [0.7–2.2 million deaths/year at 95% confidence], roughly 17% of all deaths in China.

Abstract: The first Stroke Recovery and Rehabilitation Roundtable established a game changing set of new standards for stroke recovery research. Common language and definitions were required to develop an agreed framework spanning the four working groups: translation of basic science, biomarkers of stroke recovery, measurement in clinical trials and intervention development and reporting. This paper outlines the working definitions established by our group and an agreed vision for accelerating progress in stroke recovery research.

Abstract: Co-production is currently one of cornerstones of public policy reform across the globe. Inter alia, it is articulated as a valuable route to public service reform and to the planning and delivery of effective public services, a response to the democratic deficit and a route to active citizenship and active communities, and as a means by which to lever in additional resources to public service delivery. Despite these varied roles, co-production is actually poorly formulated and has become one of a series of ‘woolly-words’ in public policy. This paper presents a conceptualization of co-production that is theoretically rooted in both public management and service management theory. It argues that this is a robust starting point for the evolution of new research and knowledge about co-production and for the development of evidence-based public policymaking and implementation.

Abstract: In this paper we present a method to capture video-wide temporal information for action recognition. We postulate that a function capable of ordering the frames of a video temporally (based on the appearance) captures well the evolution of the appearance within the video. We learn such ranking functions per video via a ranking machine and use the parameters of these as a new video representation. The proposed method is easy to interpret and implement, fast to compute and effective in recognizing a wide variety of actions. We perform a large number of evaluations on datasets for generic action recognition (Hollywood2 and HMDB51), fine-grained actions (MPII- cooking activities) and gestures (Chalearn). Results show that the proposed method brings an absolute improvement of 7–10%, while being compatible with and complementary to further improvements in appearance and local motion based methods.

Abstract: Numerical optimization has played a central role in addressing key signal processing (SP) problems Highly effective methods have been developed for a large variety of SP applications such as communications, radar, filter design, and speech and image analytics, just to name a few However, optimization algorithms often entail considerable complexity, which creates a serious gap between theoretical design/analysis and real-time processing In this paper, we aim at providing a new learning-based perspective to address this challenging issue The key idea is to treat the input and output of an SP algorithm as an unknown nonlinear mapping and use a deep neural network (DNN) to approximate it If the nonlinear mapping can be learned accurately by a DNN of moderate size, then SP tasks can be performed effectively—since passing the input through a DNN only requires a small number of simple operations In our paper, we first identify a class of optimization algorithms that can be accurately approximated by a fully connected DNN Second, to demonstrate the effectiveness of the proposed approach, we apply it to approximate a popular interference management algorithm, namely, the WMMSE algorithm Extensive experiments using both synthetically generated wireless channel data and real DSL channel data have been conducted It is shown that, in practice, only a small network is sufficient to obtain high approximation accuracy, and DNNs can achieve orders of magnitude speedup in computational time compared to the state-of-the-art interference management algorithm

Abstract: The copper metabolism disorder Wilson's disease was first defined in 1912. Wilson's disease can present with hepatic and neurological deficits, including dystonia and parkinsonism. Early-onset presentations in infancy and late-onset manifestations in adults older than 70 years of age are now well recognised. Direct genetic testing for ATP7B mutations are increasingly available to confirm the clinical diagnosis of Wilson's disease, and results from biochemical and genetic prevalence studies suggest that Wilson's disease might be much more common than previously estimated. Early diagnosis of Wilson's disease is crucial to ensure that patients can be started on adequate treatment, but uncertainty remains about the best possible choice of medication. Furthermore, Wilson's disease needs to be differentiated from other conditions that also present clinically with hepatolenticular degeneration or share biochemical abnormalities with Wilson's disease, such as reduced serum ceruloplasmin concentrations. Disordered copper metabolism is also associated with other neurological conditions, including a subtype of axonal neuropathy due to ATP7A mutations and the late-onset neurodegenerative disorders Alzheimer's disease and Parkinson's disease.

Abstract: We present Meena, a multi-turn open-domain chatbot trained end-to-end on data mined and filtered from public domain social media conversations. This 2.6B parameter neural network is simply trained to minimize perplexity of the next token. We also propose a human evaluation metric called Sensibleness and Specificity Average (SSA), which captures key elements of a human-like multi-turn conversation. Our experiments show strong correlation between perplexity and SSA. The fact that the best perplexity end-to-end trained Meena scores high on SSA (72% on multi-turn evaluation) suggests that a human-level SSA of 86% is potentially within reach if we can better optimize perplexity. Additionally, the full version of Meena (with a filtering mechanism and tuned decoding) scores 79% SSA, 23% higher in absolute SSA than the existing chatbots we evaluated.

Abstract: An important problem in econometrics and marketing is to infer the causal impact that a designed market intervention has exerted on an outcome metric over time. This paper proposes to infer causal impact on the basis of a diffusion-regression state-space model that predicts the counterfactual market response in a synthetic control that would have occurred had no intervention taken place. In contrast to classical difference-in-differences schemes, state-space models make it possible to (i) infer the temporal evolution of attributable impact, (ii) incorporate empirical priors on the parameters in a fully Bayesian treatment, and (iii) flexibly accommodate multiple sources of variation, including local trends, seasonality and the time-varying influence of contemporaneous covariates. Using a Markov chain Monte Carlo algorithm for posterior inference, we illustrate the statistical properties of our approach on simulated data. We then demonstrate its practical utility by estimating the causal effect of an online advertising campaign on search-related site visits. We discuss the strengths and limitations of state-space models in enabling causal attribution in those settings where a randomised experiment is unavailable. The CausalImpact R package provides an implementation of our approach.

Abstract: Because of the interesting and multifunctional properties, recently, ZnO nanostructures are considered as excellent material for fabrication of highly sensitive and selective gas sensors. Thus, ZnO nanomaterials are widely used to fabricate efficient gas sensors for the detection of various hazardous and toxic gases. The presented review article is focusing on the recent developments of NO2 gas sensors based on ZnO nanomaterials. The review presents the general introduction of some metal oxide nanomaterials for gas sensing application and finally focusing on the structure of ZnO and its gas sensing mechanisms. Basic gas sensing characteristics such as gas response, response time, recovery time, selectivity, detection limit, stability and recyclability, etc are also discussed in this article. Further, the utilization of various ZnO nanomaterials such as nanorods, nanowires, nano-micro flowers, quantum dots, thin films and nanosheets, etc for the fabrication of NO2 gas sensors are also presented. Moreover, various factors such as NO2 concentrations, annealing temperature, ZnO morphologies and particle sizes, relative humidity, operating temperatures which are affecting the NO2 gas sensing properties are discussed in this review. Finally, the review article is concluded and future directions are presented.

Abstract: Image descriptors based on activations of Convolutional Neural Networks (CNNs) have become dominant in image retrieval due to their discriminative power, compactness of representation, and search efficiency. Training of CNNs, either from scratch or fine-tuning, requires a large amount of annotated data, where a high quality of annotation is often crucial. In this work, we propose to fine-tune CNNs for image retrieval on a large collection of unordered images in a fully automated manner. Reconstructed 3D models obtained by the state-of-the-art retrieval and structure-from-motion methods guide the selection of the training data. We show that both hard-positive and hard-negative examples, selected by exploiting the geometry and the camera positions available from the 3D models, enhance the performance of particular-object retrieval. CNN descriptor whitening discriminatively learned from the same training data outperforms commonly used PCA whitening. We propose a novel trainable Generalized-Mean (GeM) pooling layer that generalizes max and average pooling and show that it boosts retrieval performance. Applying the proposed method to the VGG network achieves state-of-the-art performance on the standard benchmarks: Oxford Buildings, Paris, and Holidays datasets.

Abstract: . The land and ocean absorb on average just over half of the anthropogenic emissions of carbon dioxide (CO2) every year. These CO2 "sinks" are modulated by climate change and variability. Here we use a suite of nine dynamic global vegetation models (DGVMs) and four ocean biogeochemical general circulation models (OBGCMs) to estimate trends driven by global and regional climate and atmospheric CO2 in land and oceanic CO2 exchanges with the atmosphere over the period 1990–2009, to attribute these trends to underlying processes in the models, and to quantify the uncertainty and level of inter-model agreement. The models were forced with reconstructed climate fields and observed global atmospheric CO2; land use and land cover changes are not included for the DGVMs. Over the period 1990–2009, the DGVMs simulate a mean global land carbon sink of −2.4 ± 0.7 Pg C yr−1 with a small significant trend of −0.06 ± 0.03 Pg C yr−2 (increasing sink). Over the more limited period 1990–2004, the ocean models simulate a mean ocean sink of −2.2 ± 0.2 Pg C yr−1 with a trend in the net C uptake that is indistinguishable from zero (−0.01 ± 0.02 Pg C yr−2). The two ocean models that extended the simulations until 2009 suggest a slightly stronger, but still small, trend of −0.02 ± 0.01 Pg C yr−2. Trends from land and ocean models compare favourably to the land greenness trends from remote sensing, atmospheric inversion results, and the residual land sink required to close the global carbon budget. Trends in the land sink are driven by increasing net primary production (NPP), whose statistically significant trend of 0.22 ± 0.08 Pg C yr−2 exceeds a significant trend in heterotrophic respiration of 0.16 ± 0.05 Pg C yr−2 – primarily as a consequence of widespread CO2 fertilisation of plant production. Most of the land-based trend in simulated net carbon uptake originates from natural ecosystems in the tropics (−0.04 ± 0.01 Pg C yr−2), with almost no trend over the northern land region, where recent warming and reduced rainfall offsets the positive impact of elevated atmospheric CO2 and changes in growing season length on carbon storage. The small uptake trend in the ocean models emerges because climate variability and change, and in particular increasing sea surface temperatures, tend to counter\-act the trend in ocean uptake driven by the increase in atmospheric CO2. Large uncertainty remains in the magnitude and sign of modelled carbon trends in several regions, as well as regarding the influence of land use and land cover changes on regional trends.

Abstract: We review lattice results related to pion, kaon, D-meson, B-meson, and nucleon physics with the aim of making them easily accessible to the nuclear and particle physics communities. More specifically, we report on the determination of the light-quark masses, the form factor $f_+(0)$ arising in the semileptonic $K \rightarrow \pi $ transition at zero momentum transfer, as well as the decay constant ratio $f_K/f_\pi $ and its consequences for the CKM matrix elements $V_{us}$ and $V_{ud}$. Furthermore, we describe the results obtained on the lattice for some of the low-energy constants of $SU(2)_L\times SU(2)_R$ and $SU(3)_L\times SU(3)_R$ Chiral Perturbation Theory. We review the determination of the $B_K$ parameter of neutral kaon mixing as well as the additional four B parameters that arise in theories of physics beyond the Standard Model. For the heavy-quark sector, we provide results for $m_c$ and $m_b$ as well as those for D- and B-meson decay constants, form factors, and mixing parameters. These are the heavy-quark quantities most relevant for the determination of CKM matrix elements and the global CKM unitarity-triangle fit. We review the status of lattice determinations of the strong coupling constant $\alpha _s$. Finally, in this review we have added a new section reviewing results for nucleon matrix elements of the axial, scalar and tensor bilinears, both isovector and flavor diagonal.

Abstract: Representation and learning of commonsense knowledge is one of the foundational problems in the quest to enable deep language understanding. This issue is particularly challenging for understanding casual and correlational relationships between events. While this topic has received a lot of interest in the NLP community, research has been hindered by the lack of a proper evaluation framework. This paper attempts to address this problem with a new framework for evaluating story understanding and script learning: the `Story Cloze Test’. This test requires a system to choose the correct ending to a four-sentence story. We created a new corpus of 50k five-sentence commonsense stories, ROCStories, to enable this evaluation. This corpus is unique in two ways: (1) it captures a rich set of causal and temporal commonsense relations between daily events, and (2) it is a high quality collection of everyday life stories that can also be used for story generation. Experimental evaluation shows that a host of baselines and state-of-the-art models based on shallow language understanding struggle to achieve a high score on the Story Cloze Test. We discuss these implications for script and story learning, and offer suggestions for deeper language understanding.

Abstract: At the end of the day, the lesson learnt was so simple... With online and offline connections, the world is a global village (McLuhan, 1962) and a butterfly flapping its wings in Asia can cause a hurricane all around the world (Lorenz, 1972). Currently, it seems that the global education system is in the middle of this hurricane. These times, where we are all witnessing developments warily, are certainly interesting and strange, but the hope is that lessons will have been learned once things hopefully return to normal. Though there were early warnings to be prepared (White, Ramirez, Smith, & Plonowski, 2010) and already ongoing interruptions to education (Briggs, 2018; GCPEA, 2018), this is the first crisis to occur on the global scale in the digital knowledge age, and there will be socio-cultural, economic, and political consequences in the wake of this crisis. In other words, the educational landscape will feel the rush of air from the butterfly’s flapping wings to the full extent.

Abstract: Implementing compact, low-power artificial neural processing systems with real-time on-line learning abilities is still an open challenge. In this paper we present a full-custom mixed-signal VLSI device with neuromorphic learning circuits that emulate the biophysics of real spiking neurons and dynamic synapses for exploring the properties of computational neuroscience models and for building brain-inspired computing systems. The proposed architecture allows the on-chip configuration of a wide range of network connectivities, including recurrent and deep networks with short-term and long-term plasticity. The device comprises 128 K analog synapse and 256 neuron circuits with biologically plausible dynamics and bi-stable spike-based plasticity mechanisms that endow it with on-line learning abilities. In addition to the analog circuits, the device comprises also asynchronous digital logic circuits for setting different synapse and neuron properties as well as different network configurations. This prototype device, fabricated using a 180 nm 1P6M CMOS process, occupies an area of 51.4 mm 2 , and consumes approximately 4 mW for typical experiments, for example involving attractor networks. Here we describe the details of the overall architecture and of the individual circuits and present experimental results that showcase its potential. By supporting a wide range of cortical-like computational modules comprising plasticity mechanisms, this device will enable the realization of intelligent autonomous systems with on-line learning capabilities.

Abstract: Importance Reported cases of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection likely underestimate the prevalence of infection in affected communities. Large-scale seroprevalence studies provide better estimates of the proportion of the population previously infected. Objective To estimate prevalence of SARS-CoV-2 antibodies in convenience samples from several geographic sites in the US. Design, Setting, and Participants This cross-sectional study performed serologic testing on a convenience sample of residual sera obtained from persons of all ages. The serum was collected from March 23 through May 12, 2020, for routine clinical testing by 2 commercial laboratory companies. Sites of collection were San Francisco Bay area, California; Connecticut; south Florida; Louisiana; Minneapolis-St Paul-St Cloud metro area, Minnesota; Missouri; New York City metro area, New York; Philadelphia metro area, Pennsylvania; Utah; and western Washington State. Exposures Infection with SARS-CoV-2. Main Outcomes and Measures The presence of antibodies to SARS-CoV-2 spike protein was estimated using an enzyme-linked immunosorbent assay, and estimates were standardized to the site populations by age and sex. Estimates were adjusted for test performance characteristics (96.0% sensitivity and 99.3% specificity). The number of infections in each site was estimated by extrapolating seroprevalence to site populations; estimated infections were compared with the number of reported coronavirus disease 2019 (COVID-19) cases as of last specimen collection date. Results Serum samples were tested from 16 025 persons, 8853 (55.2%) of whom were women; 1205 (7.5%) were 18 years or younger and 5845 (36.2%) were 65 years or older. Most specimens from each site had no evidence of antibodies to SARS-CoV-2. Adjusted estimates of the proportion of persons seroreactive to the SARS-CoV-2 spike protein antibodies ranged from 1.0% in the San Francisco Bay area (collected April 23-27) to 6.9% of persons in New York City (collected March 23-April 1). The estimated number of infections ranged from 6 to 24 times the number of reported cases; for 7 sites (Connecticut, Florida, Louisiana, Missouri, New York City metro area, Utah, and western Washington State), an estimated greater than 10 times more SARS-CoV-2 infections occurred than the number of reported cases. Conclusions and Relevance During March to early May 2020, most persons in 10 diverse geographic sites in the US had not been infected with SARS-CoV-2 virus. The estimated number of infections, however, was much greater than the number of reported cases in all sites. The findings may reflect the number of persons who had mild or no illness or who did not seek medical care or undergo testing but who still may have contributed to ongoing virus transmission in the population.

Abstract: Matrix metalloproteinases (MMPs) are zinc-containing endopeptidases with an extensive range of substrate specificities. Collectively, these enzymes are able to degrade various components of extracellular matrix (ECM) proteins. Based on their structure and substrate specificity, they can be categorized into five main subgroups, namely (1) collagenases (MMP-1, MMP-8 and MMP-13); (2) gelatinases (MMP-2 and MMP-9); (3) stromelysins (MMP-3, MMP-10 and MMP-11); (4) matrilysins (MMP-7 and MMP-26); and (5) membrane-type (MT) MMPs (MMP-14, MMP-15, and MMP-16). The alterations made to the ECM by MMPs might contribute in skin wrinkling, a characteristic of premature skin aging. In photocarcinogenesis, degradation of ECM is the initial step towards tumor cell invasion, to invade both the basement membrane and the surrounding stroma that mainly comprises fibrillar collagens. Additionally, MMPs are involved in angiogenesis, which promotes cancer cell growth and migration. In this review, we focus on the present knowledge about premature skin aging and skin cancers such as basal cell carcinoma (BCC), squamous cell carcinoma (SCC), and melanoma, with our main focus on members of the MMP family and their functions.

Abstract: DeepMind presented remarkably accurate predictions at the recent CASP14 protein structure prediction assessment conference. We explored network architectures incorporating related ideas and obtained the best performance with a three-track network in which information at the 1D sequence level, the 2D distance map level, and the 3D coordinate level is successively transformed and integrated. The three-track network produces structure predictions with accuracies approaching those of DeepMind in CASP14, enables the rapid solution of challenging X-ray crystallography and cryo-EM structure modeling problems, and provides insights into the functions of proteins of currently unknown structure. The network also enables rapid generation of accurate protein-protein complex models from sequence information alone, short circuiting traditional approaches which require modeling of individual subunits followed by docking. We make the method available to the scientific community to speed biological research.

Abstract: Empirical research inevitably includes constructing a data set by processing raw data into a form ready for statistical analysis. Data processing often involves choices among several reasonable options for excluding, transforming, and coding data. We suggest that instead of performing only one analysis, researchers could perform a multiverse analysis, which involves performing all analyses across the whole set of alternatively processed data sets corresponding to a large set of reasonable scenarios. Using an example focusing on the effect of fertility on religiosity and political attitudes, we show that analyzing a single data set can be misleading and propose a multiverse analysis as an alternative practice. A multiverse analysis offers an idea of how much the conclusions change because of arbitrary choices in data construction and gives pointers as to which choices are most consequential in the fragility of the result.

Abstract: The R package partykit provides a flexible toolkit for learning, representing, summarizing, and visualizing a wide range of tree-structured regression and classification models. The functionality encompasses: (a) basic infrastructure for representing trees (inferred by any algorithm) so that unified print/plot/predict methods are available; (b) dedicated methods for trees with constant fits in the leaves (or terminal nodes) along with suitable coercion functions to create such trees (e.g., by rpart, RWeka, PMML); (c) a reimplementation of conditional inference trees (ctree, originally provided in the party package); (d) an extended reimplementation of model-based recursive partitioning (mob, also originally in party) along with dedicated methods for trees with parametric models in the leaves. Here, a brief overview of the package and its design is given while more detailed discussions of items (a)-(d) are available in vignettes accompanying the package.

Abstract: Large-scale genomic studies have shown that half of epithelial ovarian cancers (EOCs) have alterations in genes regulating homologous recombination (HR) repair. Loss of HR accounts for the genomic instability of EOCs and for their cellular hyper-dependence on alternative poly-ADP ribose polymerase (PARP)-mediated DNA repair mechanisms. Previous studies have implicated the DNA polymerase θ (Polθ also known as POLQ, encoded by POLQ) in a pathway required for the repair of DNA double-strand breaks, referred to as the error-prone microhomology-mediated end-joining (MMEJ) pathway. Whether Polθ interacts with canonical DNA repair pathways to prevent genomic instability remains unknown. Here we report an inverse correlation between HR activity and Polθ expression in EOCs. Knockdown of Polθ in HR-proficient cells upregulates HR activity and RAD51 nucleofilament assembly, while knockdown of Polθ in HR-deficient EOCs enhances cell death. Consistent with these results, genetic inactivation of an HR gene (Fancd2) and Polq in mice results in embryonic lethality. Moreover, Polθ contains RAD51 binding motifs and it blocks RAD51-mediated recombination. Our results reveal a synthetic lethal relationship between the HR pathway and Polθ-mediated repair in EOCs, and identify Polθ as a novel druggable target for cancer therapy.