non-small cell lung cancer (nScLc) is one of the most common lung cancers worldwide. Accurate prognostic stratification of NSCLC can become an important clinical reference when designing therapeutic strategies for cancer patients. With this clinical application in mind, we developed a deep neural network (Dnn) combining heterogeneous data sources of gene expression and clinical data to accurately predict the overall survival of nScLc patients. Based on microarray data from a cohort set (614 patients), seven well-known NSCLC biomarkers were used to group patients into biomarker-and biomarker+ subgroups. then, by using a systems biology approach, prognosis relevance values (pRV) were then calculated to select eight additional novel prognostic gene biomarkers. finally, the combined 15 biomarkers along with clinical data were then used to develop an integrative DNN via bimodal learning to predict the 5-year survival status of NSCLC patients with tremendously high accuracy (AUC: 0.8163, accuracy: 75.44%). Using the capability of deep learning, we believe that our prediction can be a promising index that helps oncologists and physicians develop personalized therapy and build the foundation of precision medicine in the future.Lung cancer is the worldwide leading cause of cancer-related mortality, with non-small cell lung cancer (NSCLC) accounting for approximately 85% of all lung cancer patients 1 . The most common NSCLC subtypes are adenocarcinoma (ADC), squamous cell carcinoma (SQC), and large cell carcinoma. Although the overall 5-year survival rate of patients diagnosed with stage I ADC was 63%, nearly 35% of patients relapsed after surgery with a poor prognosis 2 . Adjuvant treatments have been considered ideal for ADC patients with the highest risk of recurrence or death to increase survival rates 3 . Therefore, prognostic stratification is crucial for categorizing patients to help doctors make decisions on therapeutic strategies.Recently, researchers have developed predictive methods based on gene expression profiles to classify lung cancer patients with distinct clinical outcomes, including relapse and overall survival 4 . Previous studies have shown the importance of biomarkers for NSCLC, such as EPCAM, HIF1A, PKM, PTK7, ALCAM, CADM1, and SLC2A1, which were used as a single biomarker for predicting prognostic condition or metastasis 5-11 . However, cancer is a systemic disease with complicated and illusive mechanisms that often involves multiple genes and cross-talk between pathways. Therefore, extending our understanding of NSCLC via the single gene biomarkers by studying the interactions between genes is essential for more accurate prognostic prediction.Machine learning algorithms are powerful tools that apply input features (biomarkers) to capture the complicated interdependencies between these features to accurately predict clinical outcomes 12 . In addition, predicting cancer prognosis can be improved by appropriately modeling the interactions between biomarkers compared with the single biomarker approach ...

No abstract

No abstract

BackgroundHepatitis B virus (HBV) infection results in complications such as cirrhosis and hepatocellular carcinoma. Suppressing viral replication in chronic HBV carriers is an effective approach to controlling disease progression. Although antiviral compounds are available, we aimed to identify host factors that have a significant effect on viral replication efficiency.Methods and FindingsWe studied a group of hepatitis B carriers by associating serum viral load with their respective HBV genomes, and observed a significant association between high patient serum viral load with a natural sequence variant within the HBV enhancer II (Enh II) regulatory region at position 1752. Using a viral fragment as an affinity binding probe, we isolated a host DNA-binding protein belonging to the class of heterogeneous nuclear ribonucleoproteins—hnRNP K—that binds to and modulates the replicative efficiency of HBV. In cell transfection studies, overexpression of hnRNP K augmented HBV replication, while gene silencing of endogenous hnRNP K carried out by small interfering RNAs resulted in a significant reduction of HBV viral load.ConclusionThe evidence presented in this study describes a wider role for hnRNP K beyond maintenance of host cellular functions and may represent a novel target for pharmacologic intervention of HBV replication.

The p21-activated kinase (PAK) is a downstream target of Rac and CDC42, members of the Ras-related Rho subfamily, that mediates signaling pathway leading to cytoskeletal reorganization. To investigate its function in Caenorhabditis elegans development, we have isolated the cDNA coding for the p21-activated kinase homologue (CePAK) from a C. elegans embryonic cDNA library. This 2.35-kilobase pair cDNA encodes a polypeptide of 572 amino acid residues, with the highly conserved N-terminal p21-binding and the C-terminal kinase domains. Similar to its mammalian and Drosophila counterparts, the CePAK protein expressed in E. coli exhibits binding activity toward GTP-bound CeRac1 and CDC42Ce. Polyclonal antibodies raised against the recombinant CePAK recognize a specific 70-kDa protein from embryonic extracts that displays CeRac1/CDC42Ce-binding and kinase activities. Immunofluorescence analysis indicates that CePAK is specifically expressed at the hypodermal cell boundaries during embryonic body elongation, which involves dramatic cytoskeletal reorganization. Interestingly, CeRac1 and CDC42Ce are found at the same location, which might point to their common involvement in hypodermal cell fusion, a crucial morphogenetic event for nematode development.

No abstract

We explore the fundamental limits of heterogeneous distributed detection in an anonymous sensor network with n sensors and a single fusion center. The fusion center collects the single observation from each of the n sensors to detect a binary parameter. The sensors are clustered into multiple groups, and different groups follow different distributions under a given hypothesis. The key challenge for the fusion center is the anonymity of sensors -although it knows the exact number of sensors and the distribution of observations in each group, it does not know which group each sensor belongs to. It is hence natural to consider it as a composite hypothesis testing problem. First, we propose an optimal test called mixture likelihood ratio test, which is a randomized threshold test based on the ratio of the uniform mixture of all the possible distributions under one hypothesis to that under the other hypothesis.Optimality is shown by first arguing that there exists an optimal test that is symmetric, that is, it does not depend on the order of observations across the sensors, and then proving that the mixture likelihood ratio test is optimal among all symmetric tests. Second, we focus on the Neyman-Pearson setting and characterize the error exponent of the worst-case type-II error probability as n tends to infinity, assuming the number of sensors in each group is proportional to n. Finally, we generalize our result to find the collection of all achievable type-I and type-II error exponents, showing that the boundary of the region can be obtained by solving a convex optimization problem.Our results elucidate the price of anonymity in heterogeneous distributed detection, and can be extended to M -ary hypothesis testing with heterogeneous observations generated according to hidden latent variables. The results are also applied to distributed detection under Byzantine attacks, which hints that the conventional approach based on simple hypothesis testing might be too pessimistic.University, 2 the observations, the empirical distribution (types) of the collected observation is a sufficient statistic. When the communication between each sensor and the fusion center is limited, for binary hypothesis testing it is asymptotically optimal to use the same local decision function at all sensors [2]. Hence, anonymity is not a critical issue for the classical (homogeneous) distributed detection problem.However, when the joint distribution of the observations is heterogeneous, that is, marginal distributions of observations vary across sensors, sensor anonymity may deteriorate the performance of distributed detection, even for binary hypothesis testing. One such example is distributed detection under Byzantine attack [3], where a fixed number of sensors are compromised by malicious attackers and report fake observations following certain distributions. Even if the fusion center is aware of the number of compromised sensors and the attacking strategy that renders worstcase detection performance (the least favorable distribution as consi...

No abstract