Search

An open-source, physics-based tropical cyclone downscaling model is developed, in order to generate a large climatology of tropical cyclones. The model is composed of three primary components: (1) a random seeding process that determines genesis, (2) an intensity-dependent beta-advection model that determines the track, and (3) a non-linear differential equation set that determines the intensification rate. The model is entirely forced by the large-scale environment. Downscaling ERA5 reanalysis data shows that the model is generally able to reproduce observed tropical cyclone climatology, such as the global seasonal cycle, genesis locations, track density, and lifetime maximum intensity distributions. Inter-annual variability in tropical cyclone count and power-dissipation is also well captured, on both basin-wide and global scales. Regional tropical cyclone hazard estimated by this model is also analyzed using return period maps and curves. In particular, the model is able to reasonably capture the observed return period curves of landfall intensity in various sub-basins around the globe. The incorporation of an intensity-dependent steering flow is shown to lead to regionally dependent changes in power dissipation and return periods. Advantages and disadvantages of this model, compared to other downscaling models, are also discussed.

Across languages, age of acquisition (AoA) is a critical psycholinguistic factor in lexical processing, reflecting the influence of learning experience. Early-acquired words tend to be processed more quickly and accurately than late-acquired words. Recently, an integrated view proposed that both the mappings between representations and the construction of semantic representations contribute to AoA effects, thus, predicting larger AoA effects for words with arbitrary mappings between representations as well as for tasks requiring greater semantic processing. We investigated how these predictions generalize to the Chinese language system that differs from alphabetic languages regarding the ease of mappings and semantic involvement in lexical processing. A cross-task investigation of differential psycholinguistic effects was conducted with large character naming and lexical decision datasets to establish the extent to which semantics is involved in the two tasks. We focused on examining the effect sizes of lexical-semantic variables and AoA, and the interaction between AoA and consistency. The results demonstrated that semantics influenced Chinese character naming more than lexical decision, which is in contrast with the findings related to English language, though, critically, AoA effects were more pronounced for character naming than for lexical decision. Additionally, an interaction between AoA and consistency was found in character naming. Our findings provide cross-linguistic evidence supporting the view of multiple origins of AoA effects in the language-processing system.

2019, 【要旨】

We present a model of unsupervised phonological lexicon discovery -- the problem of simultaneously learning phoneme-like and word-like units from acoustic input. Our model builds on earlier models of unsupervised phone-like unit discovery from acoustic data (Lee and Glass, 2012), and unsupervised symbolic lexicon discovery using the Adaptor Grammar framework (Johnson et al., 2006), integrating these earlier approaches using a probabilistic model of phonological variation. We show that the model is competitive with state-of-the-art spoken term discovery systems, and present analyses exploring the model's behavior and the kinds of linguistic structures it learns.

Recent assessments agree that tropical cyclone intensity should increase as the climate warms. Less agreement exists on the detection of recent historical trends in tropical cyclone intensity. We interpret future and recent historical trends by using the theory of potential intensity, which predicts the maximum intensity achievable by a tropical cyclone in a given local environment. Although greenhouse gas–driven warming increases potential intensity, climate model simulations suggest that aerosol cooling has largely canceled that effect over the historical record. Large natural variability complicates analysis of trends, as do poleward shifts in the latitude of maximum intensity. In the absence of strong reductions in greenhouse gas emissions, future greenhouse gas forcing of potential intensity will increasingly dominate over aerosol forcing, leading to substantially larger increases in tropical cyclone intensities.

Background: Circulating tumor cells' (CTCs) presence in patient's peripheral circulation has been recognized as a useful tool of liquid biopsy for cancer treatment. In addition to enumeration, analysis of associated protein and molecular alterations offers great potential for precision medicine. MiSelect R was invented for purification and characterization of rare cells including CTC. This study evaluates MiSelect R analytic performances and validates with clinical samples. Methods: Lung cancer cell line H2228 was spiked into 8 mL of whole blood (16 cells/ 8mL) for single-cell retrieval, subsequent RT-PCR and sequencing analysis on ALK-translocation. Breast cancer cell lines, SKBr3 (high EpCAM expressing) and MDA-MB-231 (low EpCAM expressing), were spiked into 8mL whole blood from healthy subjects at various concentrations (4-1024 cells/ 8mL) for recovery rate and linearity. Experiments with low cell spikes (32 cells/ 8mL) vs. high cell spikes (1024 cells/ 8mL) were carried out to assess the within-device precision. The assay specificity was examined in blood of healthy subjects. Further evaluation was performed in metastatic breast cancer (mBC) patients. Results: Spiked H2228 cells were retrieved with greater than 90% recovery rate and 70% purity from whole blood. Each retrieved single H2228 cell was confirmed to have ALK-EML4 translocation by RT-PCR and sequencing analysis. The detection ranges of 4 to 1024 SkBr3 and MDA-MB-231 cells per 8mL were confirmed to be linear with average recovery rate of 91% (R2 = 0.9949) and 70% (R2 = 0.9983), respectively. The within-device precision showed 9.23% and 7.82% CV for low cell spikes vs. high cell spikes. Great assay specificity was achieved with no CTC detected in healthy subjects (N = 42). 100% detection rate of CTC was achieved in untreated mBC patients and 47% for patients regardless of treatment history. Also, the HER2 score analyzed by MiSelect R in isolated CTCs was in agreement with that by IHC analysis. Nonetheless, the HER2 expression level was heterogeneous in the CTCs isolated from HER2-positive patients. Furthermore, the PD-L1 expression in CTCs was shown to be heterogeneous within and between cancer patients. Conclusion: The MiSelect R System showed excellent capability for single-cell retrieval of rare cells from whole blood. It also demonstrated outstanding analytic performances and clinical sensitivity for CTC detection as well as versatile capability for multiple-biomarker characterization. It potentially provides important cellular and molecular information at single-cell level from liquid biopsy for prognosis. Citation Format: Ju-Yu Tseng, Yen-Ru Chen, Chia-Ying Lee, Li-Fan Wu, Cheng-Hsu Wang, Shin-Hang Wang, Hui-Min Yu, Wei-Feng Fang, Mu-Yi Chen, Chwen-Cheng Chen. MiSelect R: A novel single-cell retrieval system for CTC from whole blood for biomarker analysis in cancer targeted and immunotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 5590.

Background: PD-L1 assessment on tumor tissue is considered as an important predictive marker for anti-PD-1 treatments and has been approved by FDA as companion diagnostics for immunocheckpoint antagonists. Nonetheless, it remains challenging because of the dynamic nature and heterogeneity of PD-L1 expression, and the availability of tumor tissue. Measurement of circulating tumor cells (CTC) serves a form of liquid biopsy and tumor biomarker due to its non-invasiveness and real-time assessment. Here, we analyzed and compared PD-L1 expression in CTCs and tumor cells harvested from surgical specimen from colorectal cancer (CRC) patients. Methods: Mesenteric vein blood (MVB) samples were collected into K2EDTA-containing vacationer tubes and subsequently performed CTC enumeration and PD-L1 analysis by MiSelect R system. Tumor tissues were homogenized via collagenase enzyme digestion to single cell suspension before subjected to MiSelect R analysis. Tumor cells and CTCs are defined as EpCAM+, cytokeratin+ and CD45- with intact nuclei. PD-L1 analysis was performed by anti-PD-L1 primary antibody (clone 28-8). Tumor cells, CTCs and PD-L1+ tumor cells were counted and compared among different clinicopathological parameters. Results: MVB sample were collected intraoperatively from 116 CRC patients across various stages (stage I: 24, II: 38, III:42 and IV:12). CTC presence correlated with T stage and CEA level (P = 0.018 and 0.049, respectively); furthermore, PD-L1+ CTC presence significantly correlated with clinical stage, N stage, microscopic blood vascular and lymphatic vascular invasion (P = 0.0017, 0.013, 0.014 and 0.004, respectively). The percentage of PD-L1+ tumor cells showed positive correlation between tumor tissue and CTC, and both correlated with tumor stage. Conclusion: PD-L1+ cell was detectable in both tumor tissue and CTC in CRC patients and its percentage increased as disease progressed, which might reflect escape of tumor cell from immune surveillance. The percentage of PD-L1 expression on CTCs correlated with that in paired tumor tissues. On the other hand, the expression of PD-L1 on CTC was significantly associated with blood and lymphatic vessel invasion, which might be one of the potential mechanisms for local and distant metastasis of cancer. Further investigation is warranted to better understand the clinical significances of PD-L1-expression on CTCs and its potential utility as prognostic biomarker for cancer immunotherapy. Citation Format: Ju-Yu Tseng, Chwen-Cheng Chen, Yen-Ru Chen, Chia-Ying Lee, Chun-Chi Lin, Shin-Hang Wang, Tzu-Chao Hung, Hong-Ling Wang, Yi Chung, Yen-Lun Tseng, Mu-Yi Chen, Jeng-Kai Jiang. Presence of PD-L1+ tumor cells in tumor tissue and blood significantly associated with T/N stage and lymphovascular invasion in colorectal cancer patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1589.

e23029 Background: Circulating tumor cells (CTC) play a prognostic role in patients with metastatic colorectal cancer (mCRC). Studies investigating detection of CTC in peripheral blood (PB) showed limited clinical sensitivity. This study compares clinical sensitivity of CTC analysis in mesenteric venous blood (MVB) vs PB. PD-L1 expression on isolated CTCs was also characterized in various stages of CRC patients. Methods: 8mL PB and MVB samples were collected into K2EDTA-containing vacationer tubes and subsequently performed CTC enumeration and PD-L1 analysis by MiSelect R system. CTC is defined as EpCAM+, cytokeratin+ and CD45- with intact nuclei. CTC and PD-L1+ CTC number were counted and correlated to the patients’ clinical manifestations. Results: Blood samples were collected from 75 CRC patients with various stages. In MVB, CTC detection rate was 20%, 33%, 44% and 58% for stage I, II, III, and IV respectively, and the overall detection rate is 41%. In PB, CTC only detected in sporadic cases (7%). Also, the amount of CTC detected in MVB was more abundant than that in PB ( P < 0.001) In addition, preoperative serum CEA levels were significantly associated with presence of CTC in MVB ( P= 0.011). PD-L1 expression was detected in 32% of the isolated CTCs, and was more abundant in those from late stage patients. The expression level of PD-L1 showed heterogeneity among CTCs. Conclusions: CTC enumeration in MVB could potentially increase the clinical sensitivity for CTC analysis in CRC. The CTC level in MVB significantly correlated with CEA levels which could serve as an important biomarker information for post-operative management. In additions, using MiSelect R System to determine PD-L1 expression on CTC could provide a promising approach for both patient selection or therapeutic monitoring of immunotherapies for CRC.

In dense Erd\H{o}s-R\'enyi random graphs, we are interested in the events where large numbers of a given subgraph occur. The mean behavior of subgraph counts is known, and only recently were the related large deviations results discovered. Consequently, it is natural to ask, can one develop efficient numerical schemes to estimate the probability of an Erd\H{o}s-R\'enyi graph containing an excessively large number of a fixed given subgraph? Using the large deviation principle we study an importance sampling scheme as a method to numerically compute the small probabilities of large triangle counts occurring within Erd\H{o}s-R\'enyi graphs. We show that the exponential tilt suggested directly by the large deviation principle does not always yield an optimal scheme. The exponential tilt used in the importance sampling scheme comes from a generalized class of exponential random graphs. Asymptotic optimality, a measure of the efficiency of the importance sampling scheme, is achieved by a special choice of the parameters in the exponential random graph that makes it indistinguishable from an Erd\H{o}s-R\'enyi graph conditioned to have many triangles in the large network limit. We show how this choice can be made for the conditioned Erd\H{o}s-R\'enyi graphs both in the replica symmetric phase as well as in parts of the replica breaking phase to yield asymptotically optimal numerical schemes to estimate this rare event probability., Comment: 31 pages, 4 figures, 4 tables

In dense Erd��s-R��nyi random graphs, we are interested in the events where large numbers of a given subgraph occur. The mean behavior of subgraph counts is known, and only recently were the related large deviations results discovered. Consequently, it is natural to ask, can one develop efficient numerical schemes to estimate the probability of an Erd��s-R��nyi graph containing an excessively large number of a fixed given subgraph? Using the large deviation principle we study an importance sampling scheme as a method to numerically compute the small probabilities of large triangle counts occurring within Erd��s-R��nyi graphs. We show that the exponential tilt suggested directly by the large deviation principle does not always yield an optimal scheme. The exponential tilt used in the importance sampling scheme comes from a generalized class of exponential random graphs. Asymptotic optimality, a measure of the efficiency of the importance sampling scheme, is achieved by a special choice of the parameters in the exponential random graph that makes it indistinguishable from an Erd��s-R��nyi graph conditioned to have many triangles in the large network limit. We show how this choice can be made for the conditioned Erd��s-R��nyi graphs both in the replica symmetric phase as well as in parts of the replica breaking phase to yield asymptotically optimal numerical schemes to estimate this rare event probability., 31 pages, 4 figures, 4 tables

In an experimental study of single enzyme reactions, it has been proposed that the rate constants of the enzymatic reactions fluctuate randomly, according to a given distribution. To quantify the uncertainty arising from random rate constants, it is necessary to investigate how one can simulate such a biochemical system. To do this, we will take the Gillespie's stochastic simulation algorithm for simulating the evolution of the state of a chemical system, and study a modification of the algorithm that incorporates the random rate constants. In addition to simulating the waiting time of each reaction step, the modified algorithm also involves simulating the random fluctuation of the rate constant at each reaction time. We consider the modified algorithm in a general framework, then specialize it to two contrasting physical models, one in which the fluctuations occur on a much faster time scale than the reaction step, and the other in which the fluctuations occur much more slowly. The latter case was applied to the single enzyme reaction system, using in part the Metropolis-Hastings algorithm to enact the given distribution on the random rate constants. The modified algorithm is shown to produce simulation outputs that are corroborated by the experimental results. It is hoped that this modified algorithm can subsequently be used as a tool for the estimation or calibration of parameters in the system using experimental data.

Social Networking Sites (SNSs) have been a popular phenomenon worldwide, while religion in Computer Mediated Communication (CMC) studies has gained more academic attention. Walei as a pioneering social networking site is founded by a group of Protestant Christians targeting at the youth in Taiwan, and it exemplifies religious application on the Internet. This thesis investigates what role such a website plays both for its users and against its competition, particularly in the non-Christian cultural context. Using online interviews with nine users of Walei, how the interactivity affects Christian users‟ religious identity and the impression of non-Christian users toward Christianity is explored. This thesis seeks to shed the light on the interrelation of technology and religion, and the possible influence on identity formation online.

This paper constructs a tree structure for the music rhythm using the L-system. It models the structure as an automata and derives its complexity. It also solves the complexity for the L-system. This complexity can resolve the similarity between trees. This complexity serves as a measure of psychological complexity for rhythms. It resolves the music complexity of various compositions including the Mozart effect K488. Keyword: music perception, psychological complexity, rhythm, L-system, automata, temporal associative memory, inverse problem, rewriting rule, bracketed string, tree similarity, 21 pages, 13 figures, 2 tables

This paper presents an interactive system for creating natural looking 3D hairstyles, by which users can cut, comb and perm the hair model and generate realistic hair images. The system contains three stages: modeling, styling and rendering. In the first phrase, the system produces a physical hair model using a cantilever beam simulation with collision detection. Then the styling phrase is a hair editing process, performing styling operations to change the lengths, positions and curvatures of hair strands. Seven operations are developed for styling. A special hardware rotor platform is developed to aid the hair cut process, so that manipulation by two hands that simulate the real styling is made possible. Users can interact with tools and create various convincing hairstyles. Our system together with the rotor platform increases the ease of hairstyling 60% time saving and can render natural hair images with shadow and back-lighting effects.