Search

Neuroblastoma is the most common extracranial solid tumor of childhood. While MYCN and mutant anaplastic lymphoma kinase (ALKF1174L) cooperate in tumorigenesis, how ALK contributes to tumor formation remains unclear. Here, we used a human stem cell-based model of neuroblastoma. Mis-expression of ALKF1174L and MYCN resulted in shorter latency compared to MYCN alone. MYCN tumors resembled adrenergic, while ALK/MYCN tumors resembled mesenchymal, neuroblastoma. Transcriptomic analysis revealed enrichment in focal adhesion signaling, particularly the extracellular matrix genes POSTN and FN1 in ALK/MYCN tumors. Patients with ALK-mutant tumors similarly demonstrated elevated levels of POSTN and FN1. Knockdown of POSTN, but not FN1, delayed adhesion and suppressed proliferation of ALK/MYCN tumors. Furthermore, loss of POSTN reduced ALK-dependent activation of WNT signaling. Reciprocally, inhibition of the WNT pathway reduced expression of POSTN and growth of ALK/MYCN tumor cells. Thus, ALK drives neuroblastoma in part through a feedforward loop between POSTN and WNT signaling.

210Po, 210Pb, and 234Th were determined in water columns of the East China Sea (ECS) to investigate their biogeochemical behaviors during a severe red tide event. Dissolved 210Po, 210Pb, and 234Th accounted for large fractions of the total phases. The abnormally high concentrations of dissolved 210Pb were observed. Partition behaviors of these radionuclides were influenced by particle content effect and particle composition based on distribution coefficient (Kd) vs. total suspended matter (TSM) content and Kd vs. ratios of particulate organic carbon and total suspended matter contents (POC/TSM content ratios). The peaks of mass specific activities of 210Po, 210Pb, and 234Th indicated that degraded particles could have an intensified enrichment ability for radionuclides compared with the surficial suspended matters. Fractionation factor of 210Po and 210Pb (FPo/Pb) ( > 1) and fractionation factor of 210Po and 234Th (FPo/Th) ( > 1) were much higher at algal blooming regions than that at non-blooming stations, indicating that algal blooms promoted the fractionation of 210Po against 210Pb and 234Th, and proving that 210Po exhibited a stronger affinity for biogenic particles than 210Pb and 234Th when POC content increased in the sea. POC/210Po, POC/210Pb and POC/234Th ratios (content/activity ratios) sharply decreased with depth in both algal bloom and non-bloom stations. The outbreak of algal bloom promoted the complexity of suspended particles and increased the variability of POC/tracer ratios (content/activity ratios) in the different depth of the shallow seas. More considerations should be taken to the difficulty of the selection of export interface and the suitable tracers when algal blooming occurs.

Proteome profiling by activated esters identified >9000 ligandable lysines but they are limited as covalent inhibitors due to poor hydrolytic stability. Here we report our efforts to design and discover a new series of tunable amine-reactive electrophiles (TAREs) for selective and robust labeling of lysine. The major challenges in developing selective probes for lysine are the high nucleophilicity of cysteines and poor hydrolytic stability. Our work circumvents these challenges by a unique design of the TAREs that form stable adducts with lysine and on reaction with cysteine generate another reactive electrophiles for lysine. We highlight that TAREs exhibit substantially high hydrolytic stability as compared to the activated esters and are non-cytotoxic thus have the potential to act as covalent ligands. We applied these alternative TAREs for the intracellular labeling of proteins in different cell lines, and for the selective identification of lysines in the human proteome on a global scale.

210Po, 210Pb, and 234Th were determined in water columns of the East China Sea (ECS) to investigate their biogeochemical behaviors during a severe red tide event. Dissolved 210Po, 210Pb, and 234Th accounted for large fractions of the total phases. The abnormally high concentrations of dissolved 210Pb were observed. Partition behaviors of these radionuclides were influenced by particle content effect and particle composition based on distribution coefficient (Kd) vs. total suspended matter (TSM) content and Kd vs. ratios of particulate organic carbon and total suspended matter contents (POC/TSM content ratios). The peaks of mass specific activities of 210Po, 210Pb, and 234Th indicated that degraded particles could have an intensified enrichment ability for radionuclides compared with the surficial suspended matters. Fractionation factor of 210Po and 210Pb (FPo/Pb) ( > 1) and fractionation factor of 210Po and 234Th (FPo/Th) ( > 1) were much higher at algal blooming regions than that at non-blooming stations, indicating that algal blooms promoted the fractionation of 210Po against 210Pb and 234Th, and proving that 210Po exhibited a stronger affinity for biogenic particles than 210Pb and 234Th when POC content increased in the sea. POC/210Po, POC/210Pb and POC/234Th ratios (content/activity ratios) sharply decreased with depth in both algal bloom and non-bloom stations. The outbreak of algal bloom promoted the complexity of suspended particles and increased the variability of POC/tracer ratios (content/activity ratios) in the different depth of the shallow seas. More considerations should be taken to the difficulty of the selection of export interface and the suitable tracers when algal blooming occurs.

The article is concerned with the problem of multi-step financial time series forecasting of Foreign Exchange (FX) rates. To address this problem, we introduce a regression network termed RegPred Net. The exchange rate to forecast is treated as a stochastic process. It is assumed to follow a generalization of Brownian motion and the mean-reverting process referred to as the generalized Ornstein-Uhlenbeck (OU) process, with time-dependent coefficients. Using past observed values of the input time series, these coefficients can be regressed online by the cells of the first half of the network (Reg). The regressed coefficients depend only on - but are very sensitive to - a small number of hyperparameters required to be set by a global optimization procedure for which, Bayesian optimization is an adequate heuristic. Thanks to its multi-layered architecture, the second half of the regression network (Pred) can project time-dependent values for the OU process coefficients and generate realistic trajectories of the time series. Predictions can be easily derived in the form of expected values estimated by averaging values obtained by Monte Carlo simulation. The forecasting accuracy on a 100 days horizon is evaluated for several of the most important FX rates such as EUR/USD, EUR/CNY, and EUR/GBP. Our experimental results show that the RegPred Net significantly outperforms ARMA, ARIMA, LSTMs, and Autoencoder-LSTM models in terms of metrics measuring the absolute error (RMSE) and correlation between predicted and actual values (Pearson R, R-squared, MDA). Compared to black-box deep learning models such as LSTM, RegPred Net has better interpretability, simpler structure, and fewer parameters.

Community Question Answering (CQA) has gained increasing popularity in many domains. Mortgage is a complex and dynamic industry, and a flexible and efficient CQA platform can potentially enhance the quality of service for mortgage officers significantly. We have built a dynamic CQA platform with a state of the art semantic search engine based on recent Natural Language Processing (NLP) techniques to dynamically and collectively capture and transfer the maturity and tribal knowledge of the more experienced workforce to less experienced ones. The search engine allows for both keyword and natural language queries and is based on a fine-tuned domain-adapted Sentence-BERT encoder linearly composed with a TF-IDF vectorizer, and reciprocal-rank fused with a BM25 vectorizer. Domain adaptation and fine-tuning is based on publicly available mortgage corpora. Evaluation is performed on an internally annotated dataset using standard information retrieval metrics such as normalized discounted cumulative gain (nDCG), precision/recall at n, mean reciprocal rank, and mean average precision (MAP). The results indicate that our hybrid, fine-tuned, domain-adapted search engine is a more effective approach in responding to the information needs of our mortgage officers compared to traditional search techniques. We aim to publish the internally-annotated evaluation and training datasets in the near future., Comment: Presented on Document Intelligence Workshop at KDD (2021). Also available at https://document-intelligence.github.io/DI-2021/files/di-2021_final_17.pdf

YunYu Lin,1,* Yan Lei,2,* LinWei Li,2 Xiaoxing Su,2 Qiqi Tian,1 Wendy Wu2 1Respiratory Medicine, NingHai First Hospital, NingBo, 315600, People’s Republic of China; 2Berry Oncology Institutes, Berry Oncology Corporation, Fuzhou, 350200, People’s Republic of China*These authors contributed equally to this workCorrespondence: Qiqi Tian; Wendy Wu, Email 252446462@qq.com; wujy3261@berryoncology.comObjective: Previous studies have shown that fusion partners have a potential role in influencing different tumorigenic abilities of ROS1 fusion variants, as well as potential differential responses to crizotinib. Therefore, it is important to accurately identify the type of ROS1 rearrangement in NSCLC for clinical treatment selection.Materials and Methods: Deep-coverage targeting solid tumor 31 cancer-related genes panel was used to capture DNA-based NGS information to detect gene fusion. RNA fusion panel based on hybrid capture sequencing was performed to verify gene fusions from total RNA which isolated from formalin fixed paraffin-embedded (FFPE) tissue blocks.Results: Using DNA-targeted NGS method, we identified a novel BAIAP2-ROS1 fusion in a 71-year-old non-smoking female patient with stage IVB lung adenocarcinoma. Rearrangement consisted of BAIAP2 in exon1-exon13 of chr17: q23 and ROS1 in exon35-exon43 of chr6: q22, which were further confirmed by RNA-based NGS methodology. A complete kinase domain in ROS1 fusion was preserved. The patient subsequently received crizotinib and showed significant tumor reduction until 17 months, who got benefit from targeted therapy.Conclusion: This study discovered a novel BAIAP2-ROS1 rearrangement; it provides more knowledge of ROS1 fusion in clinical personalized treatment. The good response to crizotinib therapy emphasizes the importance of DNA-based and RNA-based NGS in rare fusion identification in clinical practice.Keywords: BAIAP2-ROS1 fusion, advanced lung adenocarcinoma, sensitive to crizotinib, next

In this paper, we study the identity of textual events from different documents. While the complex nature of event identity is previously studied (Hovy et al., 2013), the case of events across documents is unclear. Prior work on cross-document event coreference has two main drawbacks. First, they restrict the annotations to a limited set of event types. Second, they insufficiently tackle the concept of event identity. Such annotation setup reduces the pool of event mentions and prevents one from considering the possibility of quasi-identity relations. We propose a dense annotation approach for cross-document event coreference, comprising a rich source of event mentions and a dense annotation effort between related document pairs. To this end, we design a new annotation workflow with careful quality control and an easy-to-use annotation interface. In addition to the links, we further collect overlapping event contexts, including time, location, and participants, to shed some light on the relation between identity decisions and context. We present an open-access dataset for cross-document event coreference, CDEC-WN, collected from English Wikinews and open-source our annotation toolkit to encourage further research on cross-document tasks., Comment: CoNLL 2021 camera-ready

Empirical natural language processing (NLP) systems in application domains (e.g., healthcare, finance, education) involve interoperation among multiple components, ranging from data ingestion, human annotation, to text retrieval, analysis, generation, and visualization. We establish a unified open-source framework to support fast development of such sophisticated NLP workflows in a composable manner. The framework introduces a uniform data representation to encode heterogeneous results by a wide range of NLP tasks. It offers a large repository of processors for NLP tasks, visualization, and annotation, which can be easily assembled with full interoperability under the unified representation. The highly extensible framework allows plugging in custom processors from external off-the-shelf NLP and deep learning libraries. The whole framework is delivered through two modularized yet integratable open-source projects, namely Forte (for workflow infrastructure and NLP function processors) and Stave (for user interaction, visualization, and annotation)., Comment: 8 pages, 4 figures, EMNLP 2020

Gait as a biometric property for person identification plays a key role in video surveillance and security applications. In gait recognition, normally, gait feature such as Gait Energy Image (GEI) is extracted from one full gait cycle. However in many circumstances, such a full gait cycle might not be available due to occlusion. Thus, the GEI is not complete giving rise to a degrading in gait-based person identification rate. In this paper, we address this issue by proposing a novel method to identify individuals from gait feature when a few (or even single) frame(s) is available. To do so, we propose a deep learning-based approach to transform incomplete GEI to the corresponding complete GEI obtained from a full gait cycle. More precisely, this transformation is done gradually by training several auto encoders independently and then combining these as a uniform model. Experimental results on two public gait datasets, namely OULP and Casia-B demonstrate the validity of the proposed method in dealing with very incomplete gait cycles.

The flavour changing neutral current decays can be interesting probes for searching for New Physics. Angular distributions of the decay $\mathrm{B}^0 \to \mathrm{K}^{*0} \mu^ +\mu^-$ are studied using a sample of proton-proton collisions at $\sqrt{s} = 8~\mathrm{TeV}$ collected with the CMS detector at the LHC, corresponding to an integrated luminosity of $20.5~\mathrm{fb}^{-1}$. An angular analysis is performed to determine $P_1$ and $P_5'$, where $P_5'$ is of particular interest due to recent measurements that indicate a potential discrepancy with the standard model. Based on a sample of 1397 signal events, $P_1$ and $P_5'$ angular parameters are determined as a function of the dimuon invariant mass squared. The measurements are in agreement with standard model predictions., Comment: 7 pages, 3 figures, LHCP2017

The flavour changing neutral current decays can be interesting probes for searching for New Physics. Angular distributions of the decay $\mathrm{B}^0 \to \mathrm{K}^{*0} \mu^ +\mu^-$ are studied using a sample of proton-proton collisions at $\sqrt{s} = 8~\mathrm{TeV}$ collected with the CMS detector at the LHC, corresponding to an integrated luminosity of $20.5~\mathrm{fb}^{-1}$. An angular analysis is performed to determine $P_1$ and $P_5'$, where $P_5'$ is of particular interest due to recent measurements that indicate a potential discrepancy with the standard model. Based on a sample of 1397 signal events, $P_1$ and $P_5'$ angular parameters are determined as a function of the dimuon invariant mass squared. The measurements are in agreement with standard model predictions., Comment: 7 pages, 3 figures, LHCP2017

In this paper, we introduce \textsc{Yedda}, a lightweight but efficient and comprehensive open-source tool for text span annotation. \textsc{Yedda} provides a systematic solution for text span annotation, ranging from collaborative user annotation to administrator evaluation and analysis. It overcomes the low efficiency of traditional text annotation tools by annotating entities through both command line and shortcut keys, which are configurable with custom labels. \textsc{Yedda} also gives intelligent recommendations by learning the up-to-date annotated text. An administrator client is developed to evaluate annotation quality of multiple annotators and generate detailed comparison report for each annotator pair. Experiments show that the proposed system can reduce the annotation time by half compared with existing annotation tools. And the annotation time can be further compressed by 16.47\% through intelligent recommendation., Comment: Accepted by ACL 2018 as demonstration paper

The contractile function of the heart requires the release of Ca2+ from intracellular Ca2+ stores in the sarcoplasmic reticulum (SR) of cardiac muscle cells. The efficacy of Ca2+ release depends on the amount of Ca2+ loaded into the Ca2+ store and the way in which this 'Ca2+ load' influences the activity of the cardiac ryanodine receptor Ca2+ release channel (RyR2). The effects of the Ca2+ load on Ca2+ release through RyR2 are facilitated by: (i) the sensitivity of RyR2 itself to luminal Ca2+ concentrations; and (ii) interactions between the cardiac Ca2+-binding protein calsequestrin (CSQ) 2 and RyR2, transmitted through the 'anchoring' proteins junctin and/or triadin. Mutations in RyR2 are linked to catecholaminergic polymorphic ventricular tachycardia (CPVT) and sudden cardiac death. The tachycardia is associated with changes in the sensitivity of RyR2 to luminal Ca2+. Triadin-, junctin- or CSQ-null animals survive, but their longevity and ability to tolerate stress is compromised. These studies reveal the importance of the proteins in normal muscle function, but do not reveal the molecular nature of their functional interactions, which must be defined before changes in the proteins leading to CPVT and heart disease can be understood. Herein, we discuss known interactions between the RyR, triadin, junctin and CSQ with emphasis on the cardiac isoforms of the proteins. Where there is little known about the cardiac isoforms, we discuss evidence from skeletal isoforms.