Search

UNLABELLED: The comprehensive genomic analysis of the head and neck squamous cell carcinoma (HNSCC) oncogenome revealed the frequent loss of p16INK4A (CDKN2A) and amplification of cyclin D1 genes in most human papillomavirus-negative HNSCC lesions. However, cyclin-dependent kinase 4 and 6 (CDK4/6) inhibitors have shown modest effects in the clinic. The aberrant activation of the PI3K/mTOR pathway is highly prevalent in HNSCC, and recent clinical trials have shown promising clinical efficacy of mTOR inhibitors (mTORi) in the neoadjuvant and adjuvant settings but not in patients with advanced HNSCC. By implementing a kinome-wide CRISPR/Cas9 screen, we identified cell-cycle inhibition as a synthetic lethal target for mTORis. A combination of mTORi and palbociclib, a CDK4/6-specific inhibitor, showed strong synergism in HNSCC-derived cells in vitro and in vivo. Remarkably, we found that an adaptive increase in cyclin E1 (CCNE1) expression upon palbociclib treatment underlies the rapid acquired resistance to this CDK4/6 inhibitor. Mechanistically, mTORi inhibits the formation of eIF4G-CCNE1 mRNA complexes, with the consequent reduction in mRNA translation and CCNE1 protein expression. Our findings suggest that mTORi reverts the adaptive resistance to palbociclib. This provides a multimodal therapeutic option for HNSCC by cotargeting mTOR and CDK4/6, which in turn may halt the emergence of palbociclib resistance. SIGNIFICANCE: A kinome-wide CRISPR/Cas9 screen identified cell-cycle inhibition as a synthetic lethal target of mTORis. A combination of mTORi and palbociclib, a CDK4/6-specific inhibitor, showed strong synergistic effects in HNSCC. Mechanistically, mTORis inhibited palbociclib-induced increase in CCNE1.

Recently, large language models (LLMs) have notably positioned them as capable tools for addressing complex optimization challenges. Despite this recognition, a predominant limitation of existing LLM-based optimization methods is their struggle to capture the relationships among decision variables when relying exclusively on numerical text prompts, especially in high-dimensional problems. Keeping this in mind, we first propose to enhance the optimization performance using multimodal LLM capable of processing both textual and visual prompts for deeper insights of the processed optimization problem. This integration allows for a more comprehensive understanding of optimization problems, akin to human cognitive processes. We have developed a multimodal LLM-based optimization framework that simulates human problem-solving workflows, thereby offering a more nuanced and effective analysis. The efficacy of this method is evaluated through extensive empirical studies focused on a well-known combinatorial optimization problem, i.e., capacitated vehicle routing problem. The results are compared against those obtained from the LLM-based optimization algorithms that rely solely on textual prompts, demonstrating the significant advantages of our multimodal approach., Comment: 8pages,3 figures, 2 tables

Large language models (LLMs) have not only revolutionized natural language processing but also extended their prowess to various domains, marking a significant stride towards artificial general intelligence. The interplay between LLMs and evolutionary algorithms (EAs), despite differing in objectives and methodologies, share a common pursuit of applicability in complex problems. Meanwhile, EA can provide an optimization framework for LLM's further enhancement under black-box settings, empowering LLM with flexible global search capacities. On the other hand, the abundant domain knowledge inherent in LLMs could enable EA to conduct more intelligent searches. Furthermore, the text processing and generative capabilities of LLMs would aid in deploying EAs across a wide range of tasks. Based on these complementary advantages, this paper provides a thorough review and a forward-looking roadmap, categorizing the reciprocal inspiration into two main avenues: LLM-enhanced EA and EA-enhanced LLM. Some integrated synergy methods are further introduced to exemplify the complementarity between LLMs and EAs in diverse scenarios, including code generation, software engineering, neural architecture search, and various generation tasks. As the first comprehensive review focused on the EA research in the era of LLMs, this paper provides a foundational stepping stone for understanding the collaborative potential of LLMs and EAs. The identified challenges and future directions offer guidance for researchers and practitioners to unlock the full potential of this innovative collaboration in propelling advancements in optimization and artificial intelligence. We have created a GitHub repository to index the relevant papers: https://github.com/wuxingyu-ai/LLM4EC., Comment: evolutionary algorithm (EA), large language model (LLM), optimization problem, prompt engineering, algorithm generation, neural architecture search

In the algorithm selection research, the discussion surrounding algorithm features has been significantly overshadowed by the emphasis on problem features. Although a few empirical studies have yielded evidence regarding the effectiveness of algorithm features, the potential benefits of incorporating algorithm features into algorithm selection models and their suitability for different scenarios remain unclear. In this paper, we address this gap by proposing the first provable guarantee for algorithm selection based on algorithm features, taking a generalization perspective. We analyze the benefits and costs associated with algorithm features and investigate how the generalization error is affected by different factors. Specifically, we examine adaptive and predefined algorithm features under transductive and inductive learning paradigms, respectively, and derive upper bounds for the generalization error based on their model's Rademacher complexity. Our theoretical findings not only provide tight upper bounds, but also offer analytical insights into the impact of various factors, such as the training scale of problem instances and candidate algorithms, model parameters, feature values, and distributional differences between the training and test data. Notably, we demonstrate how models will benefit from algorithm features in complex scenarios involving many algorithms, and proves the positive correlation between generalization error bound and $\chi^2$-divergence of distributions.

Causality reveals fundamental principles behind data distributions in real-world scenarios, and the capability of large language models (LLMs) to understand causality directly impacts their efficacy across explaining outputs, adapting to new evidence, and generating counterfactuals. With the proliferation of LLMs, the evaluation of this capacity is increasingly garnering attention. However, the absence of a comprehensive benchmark has rendered existing evaluation studies being straightforward, undiversified, and homogeneous. To address these challenges, this paper proposes a comprehensive benchmark, namely CausalBench, to evaluate the causality understanding capabilities of LLMs. Originating from the causal research community, CausalBench encompasses three causal learning-related tasks, which facilitate a convenient comparison of LLMs' performance with classic causal learning algorithms. Meanwhile, causal networks of varying scales and densities are integrated in CausalBench, to explore the upper limits of LLMs' capabilities across task scenarios of varying difficulty. Notably, background knowledge and structured data are also incorporated into CausalBench to thoroughly unlock the underlying potential of LLMs for long-text comprehension and prior information utilization. Based on CausalBench, this paper evaluates nineteen leading LLMs and unveils insightful conclusions in diverse aspects. Firstly, we present the strengths and weaknesses of LLMs and quantitatively explore the upper limits of their capabilities across various scenarios. Meanwhile, we further discern the adaptability and abilities of LLMs to specific structural networks and complex chain of thought structures. Moreover, this paper quantitatively presents the differences across diverse information sources and uncovers the gap between LLMs' capabilities in causal understanding within textual contexts and numerical domains.

Large language models (LLMs) have gained widespread popularity and demonstrated exceptional performance not only in natural language processing (NLP) tasks but also in non-linguistic domains. Their potential as artificial general intelligence extends beyond NLP, showcasing promising capabilities in diverse optimization scenarios. Despite this rising trend, whether the integration of LLMs into these black-box optimization problems is genuinely beneficial remains unexplored. This paper endeavors to tackle this issue by offering deeper insights into the potential of LLMs in optimization tasks through a comprehensive investigation. Our approach involves a comprehensive evaluation, covering both discrete and continuous optimization problems, aiming to assess the efficacy and distinctive characteristics that LLMs bring to the realm of optimization. Our findings reveal both the limitations and advantages of LLMs in optimization. On one hand, despite consuming the significant power required to run the model, LLMs exhibit subpar performance and lack desirable properties in pure numerical tasks, primarily due to a mismatch between the problem domain and their processing capabilities. On the other hand, although LLMs may not be ideal for traditional numerical optimization, their potential in broader optimization contexts remains promising. LLMs exhibit the ability to solve problems in non-numerical domains and can leverage heuristics from the prompt to enhance their performance. To the best of our knowledge, this work presents the first systematic evaluation of LLMs for numerical optimization, offering a progressive, wide-coverage, and behavioral analysis. Our findings pave the way for a deeper understanding of LLMs' role in optimization and guide future application in diverse scenarios for LLMs.

Delivery Time Estimation (DTE) is a crucial component of the e-commerce supply chain that predicts delivery time based on merchant information, sending address, receiving address, and payment time. Accurate DTE can boost platform revenue and reduce customer complaints and refunds. However, the imbalanced nature of industrial data impedes previous models from reaching satisfactory prediction performance. Although imbalanced regression methods can be applied to the DTE task, we experimentally find that they improve the prediction performance of low-shot data samples at the sacrifice of overall performance. To address the issue, we propose a novel Dual Graph Multitask framework for imbalanced Delivery Time Estimation (DGM-DTE). Our framework first classifies package delivery time as head and tail data. Then, a dual graph-based model is utilized to learn representations of the two categories of data. In particular, DGM-DTE re-weights the embedding of tail data by estimating its kernel density. We fuse two graph-based representations to capture both high- and low-shot data representations. Experiments on real-world Taobao logistics datasets demonstrate the superior performance of DGM-DTE compared to baselines., Comment: Accepted by DASFAA 2023 Industry Track

Algorithm selection, a critical process of automated machine learning, aims to identify the most suitable algorithm for solving a specific problem prior to execution. Mainstream algorithm selection techniques heavily rely on problem features, while the role of algorithm features remains largely unexplored. Due to the intrinsic complexity of algorithms, effective methods for universally extracting algorithm information are lacking. This paper takes a significant step towards bridging this gap by introducing Large Language Models (LLMs) into algorithm selection for the first time. By comprehending the code text, LLM not only captures the structural and semantic aspects of the algorithm, but also demonstrates contextual awareness and library function understanding. The high-dimensional algorithm representation extracted by LLM, after undergoing a feature selection module, is combined with the problem representation and passed to the similarity calculation module. The selected algorithm is determined by the matching degree between a given problem and different algorithms. Extensive experiments validate the performance superiority of the proposed model and the efficacy of each key module. Furthermore, we present a theoretical upper bound on model complexity, showcasing the influence of algorithm representation and feature selection modules. This provides valuable theoretical guidance for the practical implementation of our method., Comment: Accepted by IJCAI 2024

Providing accurate estimated time of package delivery on users' purchasing pages for e-commerce platforms is of great importance to their purchasing decisions and post-purchase experiences. Although this problem shares some common issues with the conventional estimated time of arrival (ETA), it is more challenging with the following aspects: 1) Inductive inference. Models are required to predict ETA for orders with unseen retailers and addresses; 2) High-order interaction of order semantic information. Apart from the spatio-temporal features, the estimated time also varies greatly with other factors, such as the packaging efficiency of retailers, as well as the high-order interaction of these factors. In this paper, we propose an inductive graph transformer (IGT) that leverages raw feature information and structural graph data to estimate package delivery time. Different from previous graph transformer architectures, IGT adopts a decoupled pipeline and trains transformer as a regression function that can capture the multiplex information from both raw feature and dense embeddings encoded by a graph neural network (GNN). In addition, we further simplify the GNN structure by removing its non-linear activation and the learnable linear transformation matrix. The reduced parameter search space and linear information propagation in the simplified GNN enable the IGT to be applied in large-scale industrial scenarios. Experiments on real-world logistics datasets show that our proposed model can significantly outperform the state-of-the-art methods on estimation of delivery time. The source code is available at: https://github.com/enoche/IGT-WSDM23., Comment: 9 pages, accepted to WSDM 2023

An autonomous Artificial Internet of Things (AIoT) system for elderly dementia patients monitoring in a smart home is presented. The system mainly implements two functions based on the activity inference of the sensor data, which are real time abnormal activity monitoring and trend prediction of disease related activities. Specifically, CASAS dataset is employed to train a Random Forest (RF) model for activity inference. Then, another RF model trained by the output data of activity inference is used for abnormal activity monitoring. Particularly, RF is chosen for these tasks because of its balanced trade offs between accuracy, time efficiency, flexibility, and interpretability. Moreover, Long Short Term Memory (LSTM) is utilised to forecast the disease related activity trend of a patient. Consequently, the accuracy of two RF classifiers designed for activity inference and abnormal activity detection is greater than 99 percent and 94 percent, respectively. Furthermore, using the duration of sleep as an example, the LSTM model achieves accurate and evident future trends prediction.

Metformin may reduce the progression of head and neck squamous cell carcinoma (HNSCC); however, whether metformin acts by altering the host metabolism or targets cancer-initiating cells remains poorly understood. This gap in knowledge has prevented the stratification of patient populations who are most likely to benefit from metformin treatment. Here, we explored whether metformin acts directly on HNSCC cells to inhibit aberrant cell growth. To investigate the tumor cell autonomous effects of metformin, we engineered representative HPV- and HPV+ HNSCC cells harboring typical genetic alternations to express the yeast mitochondrial NADH dehydrogenase (NDI1) protein, which is insensitive to metformin. NDI1 expression rescued the inhibitory effects of metformin on mitochondrial complex I, abolished the ability of metformin to activate AMP-activated protein kinase, and inhibited mTOR signaling both in vitro and in vivo, and was sufficient to render metformin ineffective to prevent HNSCC tumor growth. This experimental system provided an opportunity to identify metformin-regulated transcriptional programs linked to cancer cell growth inhibition in the tumor microenvironment. Remarkably, computational analysis of the metformin-induced transcriptome revealed that metformin downregulated gene expression signatures associated with cancer stemness and epithelial-mesenchymal transition, concomitant with increased expression of squamous differentiation genes. These findings support that metformin may act directly on cancer-initiating cells to prevent their progression to HNSCC, which may inform the selection of patients at risk of developing HNSCC in future early-stage clinical trials. SIGNIFICANCE: Metformin's ability to directly target HNSCC-initiating cells instead of exerting cancer preventive activity based solely on its systemic effects may inform the selection of patients in future precision prevention trials.

BACKGROUNDThe aberrant activation of the PI3K/mTOR signaling circuitry is one of the most frequently dysregulated signaling events in head and neck squamous cell carcinoma (HNSCC). Here, we conducted a single-arm, open-label phase IIa clinical trial in individuals with oral premalignant lesions (OPLs) to explore the potential of metformin to target PI3K/mTOR signaling for HNSCC prevention.METHODSIndividuals with OPLs, but who were otherwise healthy and without diabetes, underwent pretreatment and posttreatment clinical exam and biopsy. Participants received metformin for 12 weeks (week 1, 500 mg; week 2, 1000 mg; weeks 3-12, 2000 mg daily). Pretreatment and posttreatment biopsies, saliva, and blood were obtained for biomarker analysis, including IHC assessment of mTOR signaling and exome sequencing.RESULTSTwenty-three participants were evaluable for response. The clinical response rate (defined as a ≥50% reduction in lesion size) was 17%. Although lower than the proposed threshold for favorable clinical response, the histological response rate (improvement in histological grade) was 60%, including 17% complete responses and 43% partial responses. Logistic regression analysis revealed that when compared with never smokers, current and former smokers had statistically significantly increased histological responses (P = 0.016). Remarkably, a significant correlation existed between decreased mTOR activity (pS6 IHC staining) in the basal epithelial layers of OPLs and the histological (P = 0.04) and clinical (P = 0.01) responses.CONCLUSIONTo our knowledge this is the first phase II trial of metformin in individuals with OPLs, providing evidence that metformin administration results in encouraging histological responses and mTOR pathway modulation, thus supporting its further investigation as a chemopreventive agent.TRIAL REGISTRATIONNCT02581137FUNDINGNIH contract HHSN261201200031I, grants R01DE026644 and R01DE026870.

BACKGROUNDThe aberrant activation of the PI3K/mTOR signaling circuitry is one of the most frequently dysregulated signaling events in head and neck squamous cell carcinoma (HNSCC). Here, we conducted a single-arm, open-label phase IIa clinical trial in individuals with oral premalignant lesions (OPLs) to explore the potential of metformin to target PI3K/mTOR signaling for HNSCC prevention.METHODSIndividuals with OPLs, but who were otherwise healthy and without diabetes, underwent pretreatment and posttreatment clinical exam and biopsy. Participants received metformin for 12 weeks (week 1, 500 mg; week 2, 1000 mg; weeks 3-12, 2000 mg daily). Pretreatment and posttreatment biopsies, saliva, and blood were obtained for biomarker analysis, including IHC assessment of mTOR signaling and exome sequencing.RESULTSTwenty-three participants were evaluable for response. The clinical response rate (defined as a ≥50% reduction in lesion size) was 17%. Although lower than the proposed threshold for favorable clinical response, the histological response rate (improvement in histological grade) was 60%, including 17% complete responses and 43% partial responses. Logistic regression analysis revealed that when compared with never smokers, current and former smokers had statistically significantly increased histological responses (P = 0.016). Remarkably, a significant correlation existed between decreased mTOR activity (pS6 IHC staining) in the basal epithelial layers of OPLs and the histological (P = 0.04) and clinical (P = 0.01) responses.CONCLUSIONTo our knowledge this is the first phase II trial of metformin in individuals with OPLs, providing evidence that metformin administration results in encouraging histological responses and mTOR pathway modulation, thus supporting its further investigation as a chemopreventive agent.TRIAL REGISTRATIONNCT02581137FUNDINGNIH contract HHSN261201200031I, grants R01DE026644 and R01DE026870.

BACKGROUNDThe aberrant activation of the PI3K/mTOR signaling circuitry is one of the most frequently dysregulated signaling events in head and neck squamous cell carcinoma (HNSCC). Here, we conducted a single-arm, open-label phase IIa clinical trial in individuals with oral premalignant lesions (OPLs) to explore the potential of metformin to target PI3K/mTOR signaling for HNSCC prevention.METHODSIndividuals with OPLs, but who were otherwise healthy and without diabetes, underwent pretreatment and posttreatment clinical exam and biopsy. Participants received metformin for 12 weeks (week 1, 500 mg; week 2, 1000 mg; weeks 3-12, 2000 mg daily). Pretreatment and posttreatment biopsies, saliva, and blood were obtained for biomarker analysis, including IHC assessment of mTOR signaling and exome sequencing.RESULTSTwenty-three participants were evaluable for response. The clinical response rate (defined as a ≥50% reduction in lesion size) was 17%. Although lower than the proposed threshold for favorable clinical response, the histological response rate (improvement in histological grade) was 60%, including 17% complete responses and 43% partial responses. Logistic regression analysis revealed that when compared with never smokers, current and former smokers had statistically significantly increased histological responses (P = 0.016). Remarkably, a significant correlation existed between decreased mTOR activity (pS6 IHC staining) in the basal epithelial layers of OPLs and the histological (P = 0.04) and clinical (P = 0.01) responses.CONCLUSIONTo our knowledge this is the first phase II trial of metformin in individuals with OPLs, providing evidence that metformin administration results in encouraging histological responses and mTOR pathway modulation, thus supporting its further investigation as a chemopreventive agent.TRIAL REGISTRATIONNCT02581137FUNDINGNIH contract HHSN261201200031I, grants R01DE026644 and R01DE026870.

Deep neural networks such as BERT have made great progress in relation classification. Although they can achieve good performance, it is still a question of concern whether these models recognize the directionality of relations, especially when they may lack interpretability. To explore the question, a novel evaluation task, called Relation Direction Recognition (RDR), is proposed to explore whether models learn the directionality of relations. Three metrics for RDR are introduced to measure the degree to which models recognize the directionality of relations. Several state-of-the-art models are evaluated on RDR. Experimental results on a real-world dataset indicate that there are clear gaps among them in recognizing the directionality of relations, even though these models obtain similar performance in the traditional metric (e.g. Macro-F1). Finally, some suggestions are discussed to enhance models to recognize the directionality of relations from the perspective of model design or training., Comment: 10 pages, 4 figures. accepted by IEEE Transactions on Emerging Topics in Computational Intelligence (TETCI)

Cetuximab, an mAb targeting EGFR, is a standard of care for the treatment for locally advanced or metastatic head and neck squamous cell carcinoma (HNSCC). However, despite overexpression of EGFR in more than 90% of HNSCC lesions, most patients with HNSCC fail to respond to cetuximab treatment. In addition, there are no available biomarkers to predict sensitivity or resistance to cetuximab in the clinic. Here, we sought to advance precision medicine approaches for HNSCC by identifying PI3K/mTOR signaling network-specific cetuximab resistance mechanisms. We first analyzed the frequency of genomic alterations in genes involved in the PI3K/mTOR signaling circuitry in the HNSCC TCGA dataset. Experimentally, we took advantage of CRISPR/Cas9 genome editing approaches to systematically explore the contribution of genomic alterations in each tumor suppressor gene (TSG) controlling the PI3K-mTOR pathway to cetuximab resistance in HNSCC cases that do not exhibit PIK3CA mutations. Remarkably, we found that many HNSCC cases exhibit pathway-specific gene copy number loss of multiple TSGs that normally restrain PI3K/mTOR signaling. Among them, we found that both engineered and endogenous PTEN gene deletions can mediate resistance to cetuximab. Our findings suggest that PTEN gene copy number loss, which is highly prevalent in HNSCC, may result in sustained PI3K/mTOR signaling independent of EGFR, thereby representing a promising mechanistic biomarker predictive of cetuximab resistance in this cancer type. Further prospective studies are needed to investigate the impact of PTEN loss on cetuximab efficacy in the clinic.

Causal variables in Markov boundary (MB) have been widely applied in extensive single-label tasks. While few researches focus on the causal variable discovery in multi-label data due to the complex causal relationships. Since some variables in multi-label scenario might contain causal information about multiple labels, this paper investigates the problem of multi-label causal variable discovery as well as the distinguishing between common causal variables shared by multiple labels and label-specific causal variables associated with some single labels. Considering the multiple MBs under the non-positive joint probability distribution, we explore the relationships between common causal variables and equivalent information phenomenon, and find that the solutions are influenced by equivalent information following different mechanisms with or without existence of label causality. Analyzing these mechanisms, we provide the theoretical property of common causal variables, based on which the discovery and distinguishing algorithm is designed to identify these two types of variables. Similar to single-label problem, causal variables for multiple labels also have extensive application prospects. To demonstrate this, we apply the proposed causal mechanism to multi-label feature selection and present an interpretable algorithm, which is proved to achieve the minimal redundancy and the maximum relevance. Extensive experiments demonstrate the efficacy of these contributions.

Aberrant activation of the PI3K-mTOR signaling pathway occurs in >80% of head and neck squamous cell carcinomas (HNSCC), and overreliance on this signaling circuit may in turn represent a cancer-specific vulnerability that can be exploited therapeutically. mTOR inhibitors (mTORi) promote tumor regression in genetically defined and chemically induced HNSCC animal models, and encouraging results have been recently reported. However, the mTOR-regulated targets contributing to the clinical response have not yet been identified. Here, we focused on EIF4E-BP1 (4E-BP1), a direct target of mTOR that serves as key effector for protein synthesis. A systematic analysis of genomic alterations in the PIK3CA-mTOR pathway in HNSCC revealed that 4E-BP1 is rarely mutated, but at least one 4E-BP1 gene copy is lost in over 35% of the patients with HNSCC, correlating with decreased 4E-BP1 protein expression. 4E-BP1 gene copy number loss correlated with poor disease-free and overall survival. Aligned with a tumor-suppressive role, 4e-bp1/2 knockout mice formed larger and more lesions in models of HNSCC carcinogenesis. mTORi treatment or conditional expression of a mutant 4E-BP1 that cannot be phosphorylated by mTOR was sufficient to disrupt the translation-initiation complex and prevent tumor growth. Furthermore, CRISPR/Cas9-targeted 4E-BP1 HNSCC cells resulted in reduced sensitivity to mTORi in vitro and in vivo. Overall, these findings indicate that in HNSCC, mTOR persistently restrains 4E-BP1 via phosphorylation and that mTORi can restore the tumor-suppressive function of 4E-BP1. Our findings also support 4E-BP1 expression and phosphorylation status as a mechanistic biomarker of mTORi sensitivity in patients with HNSCC. SIGNIFICANCE: These findings suggest that EIF4E-BP1 acts as a tumor suppressor in HNSCC and that 4E-BP1 dephosphorylation mediates the therapeutic response to mTORi, providing a mechanistic biomarker for future precision oncology trials.

PurposeWe studied the impact of mTOR signaling inhibition with rapamycin in head and neck squamous cell carcinoma (HNSCC) in the neoadjuvant setting. The goals were to evaluate the mTOR pathway as a therapeutic target for patients with advanced HNSCC, and the clinical safety, antitumor, and molecular activity of rapamycin administration on HNSCC.Patients and methodsPatients with untreated stage II-IVA HNSCC received rapamycin for 21 days (day 1, 15 mg; days 2-12, 5 mg) prior to definitive treatment with surgery or chemoradiation. Treatment responses were assessed clinically and radiographically with CT and FDG-PET. Pre- and posttreatment biopsies and blood were obtained for toxicity, immune monitoring, and IHC assessment of mTOR signaling, as well as exome sequencing.ResultsSixteen patients (eight oral cavity, eight oropharyngeal) completed rapamycin and definitive treatment. Half of patients were p16 positive. One patient had a pathologic complete response and four (25%) patients met RECIST criteria for response (1 CR, 3 PR, 12 SD). Treatment was well tolerated with no grade 4 or unexpected toxicities. No significant immune suppression was observed. Downstream mTOR signaling was downregulated in tumor tissues as measured by phosphorylation of S6 (P < 0.0001), AKT (P < 0.0001), and 4EBP (P = 0.0361), with a significant compensatory increase in phosphorylated ERK in most patients (P < 0.001). Ki67 was reduced in tumor biopsies in all patients (P = 0.013).ConclusionsRapamycin treatment was well tolerated, reduced mTOR signaling and tumor growth, and resulted in significant clinical responses despite the brief treatment duration, thus supporting the potential role of mTOR inhibitors in treatment regimens for HNSCC.

In this paper we present an end-to-end framework for addressing the problem of dynamic pricing (DP) on E-commerce platform using methods based on deep reinforcement learning (DRL). By using four groups of different business data to represent the states of each time period, we model the dynamic pricing problem as a Markov Decision Process (MDP). Compared with the state-of-the-art DRL-based dynamic pricing algorithms, our approaches make the following three contributions. First, we extend the discrete set problem to the continuous price set. Second, instead of using revenue as the reward function directly, we define a new function named difference of revenue conversion rates (DRCR). Third, the cold-start problem of MDP is tackled by pre-training and evaluation using some carefully chosen historical sales data. Our approaches are evaluated by both offline evaluation method using real dataset of Alibaba Inc., and online field experiments starting from July 2018 with thousands of items, lasting for months on Tmall.com. To our knowledge, there is no other DP field experiment using DRL before. Field experiment results suggest that DRCR is a more appropriate reward function than revenue, which is widely used by current literature. Also, continuous price sets have better performance than discrete sets and our approaches significantly outperformed the manual pricing by operation experts., Comment: 9 pages, 7 figures

Recent studies have characterized the extensive somatic alterations that arise during cancer. However, the somatic evolution of a tumor may be significantly affected by inherited polymorphisms carried in the germline. Here, we analyze genomic data for 5,954 tumors to reveal and systematically validate 412 genetic interactions between germline polymorphisms and major somatic events, including tumor formation in specific tissues and alteration of specific cancer genes. Among germline-somatic interactions, we found germline variants in RBFOX1 that increased incidence of SF3B1 somatic mutation by 8-fold via functional alterations in RNA splicing. Similarly, 19p13.3 variants were associated with a 4-fold increased likelihood of somatic mutations in PTEN. In support of this association, we found that PTEN knockdown sensitizes the MTOR pathway to high expression of the 19p13.3 gene GNA11 Finally, we observed that stratifying patients by germline polymorphisms exposed distinct somatic mutation landscapes, implicating new cancer genes. This study creates a validated resource of inherited variants that govern where and how cancer develops, opening avenues for prevention research.Significance: This study systematically identifies germline variants that directly affect tumor evolution, either by dramatically increasing alteration frequency of specific cancer genes or by influencing the site where a tumor develops. Cancer Discovery; 7(4); 410-23. ©2017 AACR.See related commentary by Geeleher and Huang, p. 354This article is highlighted in the In This Issue feature, p. 339.

Motion trajectory recognition is important for characterizing the moving property of an object. The speed and accuracy of trajectory recognition rely on a compact and discriminative feature representation, and the situations of varying rotation, scaling and translation has to be specially considered. In this paper we propose a novel feature extraction method for trajectories. Firstly a trajectory is represented by a proposed point context, which is a rotation-scale-translation (RST) invariant shape descriptor with a flexible tradeoff between computational complexity and discrimination, yet we prove that it is a complete shape descriptor. Secondly, the shape context is nonlinearly mapped to a subspace by kernel nonparametric discriminant analysis (KNDA) to get a compact feature representation, and thus a trajectory is projected to a single point in a low-dimensional feature space. Experimental results show that, the proposed trajectory feature shows encouraging improvement than state-of-art methods., Comment: 11 pages, 10 figures