Cytoscape is an open source software project for integrating biomolecular interaction networks with high-throughput expression data and other molecular states into a unified conceptual framework. Although applicable to any system of molecular components and interactions, Cytoscape is most powerful when used in conjunction with large databases of protein-protein, protein-DNA, and genetic interactions that are increasingly available for humans and model organisms. Cytoscape's software Core provides basic functionality to layout and query the network; to visually integrate the network with expression profiles, phenotypes, and other molecular states; and to link the network to databases of functional annotations. The Core is extensible through a straightforward plug-in architecture, allowing rapid development of additional computational analyses and features. Several case studies of Cytoscape plug-ins are surveyed, including a search for interaction pathways correlating with changes in gene expression, a study of protein complexes involved in cellular recovery to DNA damage, inference of a combined physical/functional interaction network for Halobacterium, and an interface to detailed stochastic/kinetic gene regulatory models.[The Cytoscape v1.1 Core runs on all major operating systems and is freely available for download from http://www.cytoscape.org/ as an open source Java application.] Such models promise to transform biological research by providing a framework to (1) systematically interrogate and experimentally verify knowledge of a pathway; (2) manage the immense complexity of hundreds or potentially thousands of cellular components and interactions; and (3) reveal emergent properties and unanticipated consequences of different pathway configurations.Typically, models are directed toward a cellular process or disease pathway of interest (Gilman and Arkin 2002) and are built by formulating existing literature as a system of differential and/or stochastic equations. However, pathway-specific models are now being supplemented with global data gathered for an entire cell or organism, by use of two complementary approaches. First, recent technological developments have made it feasible to measure pathway structure systematically, using highthroughput screens for protein-protein (Ito et al. 2001;von Mering et al. 2002), protein-DNA (Lee et al. 2002, and genetic interactions (Tong et al. 2001). To complement these data, a second set of high-throughput methods are available to characterize the molecular and cellular states induced by pathway interactions under different experimental conditions. For instance, global changes in gene expression are measured with DNA microarrays (DeRisi et al. 1997), whereas changes in protein abundance (Gygi et al. 1999), protein phosphorylation state (Zhou et al. 2001), and metabolite concentrations (Griffin et al. 2001) may be quantified with mass spectrometry, NMR, and other advanced techniques. High-throughput data pertaining to molecular interactions and states are well matched, in...

No abstract

Federated learning (FL) is a machine learning setting where many clients (e.g. mobile devices or whole organizations) collaboratively train a model under the orchestration of a central server (e.g. service provider), while keeping the training data decentralized. FL embodies the principles of focused data collection and minimization, and can mitigate many of the systemic privacy risks and costs resulting from traditional, centralized machine learning and data science approaches. Motivated by the explosive growth in FL research, this paper discusses recent advances and presents an extensive collection of open problems and challenges.

Modern mobile devices have access to a wealth of data suitable for learning models, which in turn can greatly improve the user experience on the device. For example, language models can improve speech recognition and text entry, and image models can automatically select good photos. However, this rich data is often privacy sensitive, large in quantity, or both, which may preclude logging to the data center and training there using conventional approaches. We advocate an alternative that leaves the training data distributed on the mobile devices, and learns a shared model by aggregating locally-computed updates. We term this decentralized approach Federated Learning.We present a practical method for the federated learning of deep networks based on iterative model averaging, and conduct an extensive empirical evaluation, considering five different model architectures and four datasets. These experiments demonstrate the approach is robust to the unbalanced and non-IID data distributions that are a defining characteristic of this setting. Communication costs are the principal constraint, and we show a reduction in required communication rounds by 10-100× as compared to synchronized stochastic gradient descent.

No abstract