Accurate inference of absolute copy numbers beyond simple gains and losses from single-cell chromatin accessibility (scATAC-seq) data remains challenging, thereby obscuring the distinction between genetic and epigenetically driven oncogenic dependencies. Here, we present TeaCNV, a computational framework that reconstructs clonal absolute copy number profiles and tumor clonal architectures from scATAC-seq data without matched DNA baselines. Through validation both in silico and against bulk whole-genome sequencing in renal cell carcinomas, TeaCNV resolved subclonal absolute copy number profiles with less than 10% error and detected copy number variations (CNVs) with 98.6% accuracy, outperforming existing methods. Applied to six cancer types including renal, breast, pancreatic, head and neck, colorectal, and ovarian cancers, TeaCNV delineated polyclonal architectures and revealed distinct chromatin accessibility patterns driven by CNVs in key driver genes, including AKT2, ZNF217, and SOX2. By enabling absolute copy number profiling and clonal deconvolution from epigenomic assays, TeaCNV bridges critical gaps in studying oncogenic dependencies and genotype-phenotype relationships at single-cell resolution.

Cancer is characterized by aneuploidy, often resulting from chromosomal instability (CIN) due to events such as whole-genome duplication (WGD), breakage-fusion-bridge (BFB) cycles, and chromothripsis. These catastrophic events drive de novo copy number alterations (CNAs) with different chromosomal rearrangement patterns, contributing to tumor heterogeneity and evolution. The lack of methods capable of dissecting complex patterns of CNAs at single-cell resolution limits the understanding of how chromosome aneuploidy influences tumor evolution. To address this, we developed ChromoPattern, a computational framework to quantify four distinct chromosomal rearrangement patterns associated with complex CNA mechanisms based on single-cell copy number profiles. ChromoPattern effectively differentiates these patterns through simulations and real-world data. We applied ChromoPattern to single-cell DNA sequencing data from a liver cancer cell line and breast cancer patients. Our analysis identified the oscillation pattern of CNAs as a significant contributor to tumor heterogeneity and evolution. This pattern was associated with a 20% increase in the de novo CNA rate during tumor evolution and was linked to increased intratumor heterogeneity and subclonality. Further analysis using single-cell RNA sequencing data demonstrated that tumor clones with an enriched oscillation pattern display increased cell cycle activity and a stem-like phenotype. These findings underscore the significant and differential roles of chromosomal rearrangement patterns in tumor heterogeneity and evolution. The identification of the oscillation pattern as a key driver of tumor heterogeneity and evolution opens new avenues for developing targeted therapies and diagnostic tools.

Copy number alterations (CNAs) are a key characteristic of tumor development and progression. The accumulation of various CNAs during tumor development plays a critical role in driving tumor evolution. Heterogeneous clones driven by distinct CNAs have different selective advantages, leading to differential patterns of tumor evolution that are essential for developing effective cancer therapies. Recent advances in single-cell sequencing technology have enabled genome-wide copy number profiling of tumor cell populations at single-cell resolution. This has made it possible to explore the evolutionary patterns of CNAs and accurately discover the mechanisms of intra-tumor heterogeneity. Here, we propose a two-step statistical approach that distinguishes neutral, linear, branching and punctuated evolutionary patterns for a tumor cell population based on single-cell copy number profiles. We assessed our approach using a variety of simulated and real single-cell genomic and transcriptomic datasets, demonstrating its high accuracy and robustness in predicting tumor evolutionary patterns. We applied our approach to single-cell DNA sequencing data from 20 breast cancer patients and observed that punctuated evolution is the dominant evolutionary pattern in breast cancer. Similar conclusions were drawn when applying the approach to single-cell RNA sequencing data obtained from 132 various cancer patients. Moreover, we found that differential immune cell infiltration is associated with specific evolutionary patterns. The source code of our study is available at https://github.com/FangWang-SYSU/PTEM.

Dependency-aware jobs, such as the big data analytic workflows, are commonly executed on the cloud. They are compiled to directed acyclic graphs, with tasks linked in regarding the dependency. The cloud scheduler, which maintains a large number of resources, is responsible to execute tasks in parallel. To resolve the complex dependencies, Deep Reinforcement Learning (DRL) based schedulers are widely applied. However, we find that the DRL-based schedulers are vulnerable to the perturbations in the input jobs and may generate falsified decisions to benefit a particular job while delaying the others. By perturbation, we mean a slight adjustment to the job's node features or dependencies, while not changing its functionality. In this paper, we first explore the vulnerability of DRL-based schedulers to job perturbations without accessing the information of the DRL models used in the scheduler. We devise the black-box perturbation system, in which, a proxy model is trained to mimic the DRL-based scheduling policy. We show that the high-faith proxy model can help to craft effective perturbations. The DRL-based schedulers can be as high as 60% likely to be badly affected by the perturbations. Then, we investigate the solution to improve the robustness of DRL-based schedulers to such perturbations. We propose an adversarial training framework to force the neural model to adapt to the perturbation patterns during training so as to eliminate the potential damage during applications. Experiments show that the adversarial-trained scheduler is more robust, reducing the chance of being affected to 3-fold less and the potential bad effects halved.

Recently, Deep Learning (DL) is widely applied to intelligent systems equipped with resource-constraint hardware accelerators. With multiple DL applications sharing the resource, the execution model can be divided into two stages: (i) batching independent inference tasks initiated by each application, and (ii) scheduling batches to run in a time-sharing manner. The state-of-the-art DL serving systems employ the execution model by organizing sequential tasks into batches and then scheduling batches concerning their targeting deep neural network (DNN) models in a round-robin manner. However, we demonstrated that these practices fail to alleviate the slowdown of tasks, and there is a need to re-visit batching and scheduling in terms of efficiency and fairness. To this end, we formulated batching as a resource allocation problem and investigated scheduling in terms of each application's utilization on the device. Then, we proposed the fine-grained batching scheme and fairness-driven scheduling scheme for DL serving and implemented a prototype system called DyBatch. To be exact, DyBatch accomplishes efficient batching by taking into account Pareto efficiency of and envy between batches. Besides, DyBatch's fair scheduler monitors the resource utilization of all applications and assigns a batch from the application with the lowest utilization for execution first. Evaluation under various benchmarks with comparison to the baseline system Tensorflow Serving (TFS) shows the superiority of DyBatch, which achieves up to 55% reduction of slowdown, and up to 12% improvement of throughput.

A scheduler is essential for resource management in a shared computer cluster, particularly scheduling algorithms play an important role in meeting service level objectives of user applications in large scale clusters that underlie cloud computing. Traditional cluster schedulers are often based on empirical observations of patterns of jobs running on them. It is unclear how effective they are for capturing the patterns of a variety of jobs in clouds. Recent advances in Deep Reinforcement Learning (DRL) promise a new optimization framework for a scheduler to systematically address the problem. A DRL-based scheduler can extract detailed patterns from job features and the dynamics of cloud resource utilization for better scheduling decisions. However, the deep neural network models used by the scheduler might be vulnerable to adversarial attacks. There is limited research investigating the vulnerability in DRL-based schedulers. In this paper, we give a white-box attack method to show that malicious users can exploit the scheduling vulnerability to benefit certain jobs. The proposed attack method only requires minor perturbations job features to significantly change the scheduling priority of these jobs. We implement both greedy and critical path based algorithms to facilitate the attacks to a state-of-the-art DRL based scheduler called Decima. Our extensive experiments on TPC-H workloads show a 62% and 66% success rate of attacks with the two algorithms. Successful attacks achieve a 18.6% and 17.5% completion time reduction.

In recent years, Edge computing has emerged as a new paradigm that can reduce communication delays over the Internet by moving computation power from far-end cloud servers to be closer to data sources. It is natural to shift the design of cloud-based IoT applications to Edge-based ones. Activity recognition in smart homes is one of the IoT applications that can benefit significantly from such a shift. In this work, we propose an Edge-based solution for addressing the activity recognition problem in smart homes from multiple perspectives, including architecture, algorithm design and system implementation. First, the Edge computing architecture is introduced and several critical management tasks are also investigated. Second, a realization of the Edge computing system is presented by using open source software and low-cost hardware. The consistency and scalability of running jobs on Edge devices are also addressed in our approach. Last, we propose a convolutional neural network model to perform activity recognition tasks on Edge devices. Preliminary experiments are conducted to compare our model with existing machine learning methods, and the results demonstrate that the performance of our model is promising.

Nowadays, high dynamic range (HDR) images have been widely used in photography, video processing and visual reality. In order to display HDR image on the ordinary low dynamic range devices, a technique named tone-mapping is introduced to reform its intensity value and produce a low dynamic range (LDR) image while keeping high contrast and rich detail. According to the work [1–5], the operators are always divided to two kinds of tone-mapping operators: spatially varying and spatially uniform. In the paper, we propose a new spatially varying operator based on the layered iCAM06 model and contrast enhanced adaptive histogram equalization. We first decompose the HDR image into base layer and detail layer, and then apply the contrast enhanced operator to both layers. Last, we experiment the operator on a series of HDR luminance image, and get satisfactory result with improved contrast and visual sensation, as well as lowered time cost with respect the original iCAM06 operator. The results are illustrated in detail and evaluated with contrast and entropy. It shows that our operator improves the contrast and details, as well as the visual sensation of all the LDR images comparing to the original operator.

The Cancer Genome Atlas project has generated multi-dimensional and highly integrated genomic data from a large number of patient samples with detailed clinical records across many cancer types, but it remains unclear how to best integrate the massive amount of genomic data into clinical practice. We report here our methodology to build a multi-dimensional subnetwork atlas for cancer prognosis to better investigate the potential impact of multiple genetic and epigenetic (gene expression, copy number variation, microRNA expression and DNA methylation) changes on the molecular states of networks that in turn affects complex cancer survivorship. We uncover an average of 38 novel subnetworks in the protein-protein interaction network that correlate with prognosis across four prominent cancer types. The clinical utility of these subnetwork biomarkers was further evaluated by prognostic impact evaluation, functional enrichment analysis, drug target annotation, tumor stratification and independent validation. Some pathways including the dynactin, cohesion and pyruvate dehydrogenase-related subnetworks are identified as promising new targets for therapy in specific cancer types. In conclusion, this integrative analysis of existing protein interactome and cancer genomics data allows us to systematically dissect the molecular mechanisms that underlie unexpected outcomes for cancer, which could be used to better understand and predict clinical outcomes, optimize treatment and to provide new opportunities for developing therapeutics related to the subnetworks identified.

Gene expression is modulated by multiple mechanisms, including genetic and/or epigenetic regulation, and associated with the processes of cellular differentiation and morphogenesis. Single nucleotide polymorphisms (SNPs) and DNA methylation play important roles in regulating gene expression. In this study, we focused on revealing the relationship between SNPs, DNA methylation and gene expression in two human populations genome-wide through proposing four regulation patterns and developed maximum likelihood estimate models. Using simulated data with different correlation coefficients between any two traits, the power of our approach showed a favourable performance and relative stability. In all, 6733 SNP-CpG-gene pairs including 957 genes were obtained in Northern European ancestry (CEU) population. As the results showed, SNPs and DNA methylation had approximately the same effect on expression regulation of 49% genes, which was termed cooperative/antagonistic regulation pattern. Less than 30% of genes are controlled only by one of the factors (SNP/DNA methylation). The others showed SNPs that affect methylation have no consequent effects or crosstalk regulation on gene expression. Similar result was shown in Yourba (YRI) population. Specific genes were inferred using the different mechanisms of gene regulation involved in complex diseases by combining literature. This approach provides a method to comprehensively assess regulation patterns of gene expression in the whole genome.

MOTIVATION: Recent studies on human disease have revealed that aberrant interaction between proteins probably underlies a substantial number of human genetic diseases. This suggests a need to investigate disease inheritance mode using interaction, and based on which to refresh our conceptual understanding of a series of properties regarding inheritance mode of human disease. RESULTS: We observed a strong correlation between the number of protein interactions and the likelihood of a gene causing any dominant diseases or multiple dominant diseases, whereas no correlation was observed between protein interaction and the likelihood of a gene causing recessive diseases. We found that dominant diseases are more likely to be associated with disruption of important interactions. These suggest inheritance mode should be understood using protein interaction. We therefore reviewed the previous studies and refined an interaction model of inheritance mode, and then confirmed that this model is largely reasonable using new evidences. With these findings, we found that the inheritance mode of human genetic diseases can be predicted using protein interaction. By integrating the systems biology perspectives with the classical disease genetics paradigm, our study provides some new insights into genotype-phenotype correlations. CONTACT: [email protected] or [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

The rate equations for the intracavity-frequency-doubled quasi-three-level lasers are developed. By normalizing the related parameters, it is shown that the general solution to the rate equations is dependent upon four dimensionless parameters: the normalized reabsorption loss, the pump to laser-mode size ratio, the normalized pump level, and a parameter written as η shg , which is related to the ability of the nonlinear crystal to convert the fundamental to the second harmonic. By numerically solving these rate equations, a group of general curves are obtained to express the relations between the solution and the four dimensionless parameters.