Cytoscape Automation: empowering workflow-based network analysis.

Otasek D, Morris JH, Bouças J, Pico AR, Demchak B. · 2019

Read Cytoscape Automation: empowering workflow-based network analysis. by Otasek D, Morris JH, Bouças J, Pico AR, Demchak B. (2019) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc6717989.txt.

Introduction

Otasek and co-authors describe Cytoscape Automation, a bridge between Cytoscape and workflow tools. Cytoscape is a popular program for viewing and analyzing biological networks. Its point-and-click style, though, makes work hard to repeat or scale. The new system exposes Cytoscape functions to code instead.

More than 270 core functions and 34 apps can now be called through a standard web interface. Python and R libraries in tools like Jupyter wrap those calls. Published workflows already use the system. The change lets researchers script network analysis, share exact steps, and rerun the same work on new data without clicking through menus.

Catalog record

CREATOROtasek D, Morris JH, Bouças J, Pico AR, Demchak B.
YEAR2019
COLLECTIONPAPERS
FORMATTXT
FILE SIZE54 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC6717989

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Cytoscape Automation: empowering workflow-based network analysis By David Otasek, John H. Morris, Jorge Bouças, Alexander R. Pico, Barry Demchak Published in: Genome Biology DOI: 10.1186/s13059-019-1758-4 ABSTRACT Cytoscape is one of the most successful network biology analysis and visualization tools, but because of its interactive nature, its role in creating reproducible, scalable, and novel workflows has been limited. We describe Cytoscape Automation (CA), which marries Cytoscape to highly productive workflow systems, for example, Python/R in Jupyter/RStudio. We expose over 270 Cytoscape core functions and 34 Cytoscape apps as REST-callable functions with standardized JSON interfaces backed by Swagger documentation. Independent projects to create and publish Python/R native CA interface libraries have reached an advanced stage, and a number of automation workflows are already published. Electronic supplementary material The online version of this article (10.1186/s13059-019-1758-4) contains supplementary material, which is available to authorized users. INTRODUCTION As a platform for network biologic analysis, Cytoscape [1] has proven to be enormously popular, with over 17,600 downloads worldwide each month, 5000 startups each day, and over 1000 direct citations per year. Investigators can interactively explore complex *omics datasets via analysis and visualization functions provided by Cytoscape and a large and vibrant community of app contributors. However, interactive use has proven inadequate for precisely reproducing or sharing complex analyses or for scaling to high volume or production analysis. Moreover, while Cytoscape apps provide highly performant and relevant network biology functionality, the specialized programming talent and relatively long development times they require can make them uneconomical for delivering complex and evolving workflows. Finally, as an interactive tool, Cytoscape is not positioned to add value to emerging workflows that integrate one or more external data acquisition and analysis tools (e.g., Galaxy [2], Taverna [3], and libraries provided in repositories such as PyPI [4] and Bioconductor [5]). As shown in Fig. 1, Cytoscape Automation [6] is a new Cytoscape feature that addresses these issues by extending the existing CyREST [7, 8] app, which empowers bioinformaticians to create reproducible workflows expressed in robust and well-known programming languages (e.g., Python, R, Javascript) using familiar programming environments (e.g., Jupyter and RStudio). Under Cytoscape Automation, workflows can use CyREST to issue commands to Cytoscape and automation-enabled apps via the REST protocol, which encodes data as JSON documents. Both REST and JSON are already in wide use in client/server computing, are accessible from most programming languages, are immediately understood by most bioinformaticians, and are easy to learn given the massive body of relevant training materials, examples, and extant community. Fig. 1Overview of the Cytoscape Automation ecosystem. Reproducible workflows (as Python/R/Javascript or Cytoscape Command scripts) and datasets control Cytoscape through Cytoscape Automation. Results can be created either directly from Cytoscape or from Python/R/Javascript themselves This paper focuses on using Cytoscape Automation from Python and R because they are widely used and understood by bioinformaticists and because they already have well-documented repositories of bioinformatic functions that enable researchers to create reliable, flexible, and performant bioinformatic workflows quickly and easily. Our py2cytoscape [9] (for Python) and RCy3 [10] (for R) libraries provide easy access to Cytoscape and app functionality and are available in these repositories, too. Library download statistics reported by GitHub, PyPI, and Bioconductor indicate that researcher interest in Cytoscape Automation is strong—500 downloads/month for py2Cytoscape and 800 downloads/month for RCy3. Critically, Cytoscape Automation creates new standards that encourage Cytoscape core and app authors to expose Cytoscape functvionality via REST-based API calls backed by state-of-the-art documentation based on the widely used Swagger [11] documentation framework. Swagger is purpose-built to improve workflow author productivity in a REST context by presenting complete CyREST endpoint documentation, organizing endpoints by category, and assisting in workflow prototyping via an easy click-to-run web-based interface. As a result, novel network biologic workflows can now be quickly and cheaply delivered as integrations of Cytoscape functions, complex custom analyses, and best-of-breed external tools and language-specific libraries. In this paper, we explain the key features of Cytoscape Automation, including how they work, how Cytoscape app developers can make automation-enabled apps, and how workflow authors can leverage Cytoscape Automation to create and evolve their workflows. As a reference material, we provide the substantial Cytoscape Automatic Wiki [12], which contains articles on context, implementation details, FAQs, best practices, and sample scripts and apps to help workflow authors become quickly productive and help Cytoscape app authors produce new automation-capable apps or upgrade existing ones. As an illustration aid, we use the running example of Cytoscape Diffusion [13], which uses network propagation to find new nodes (e.g., genes) that are most relevant to a set of well-understood nodes. Diffusion is particularly apt because it shows how to define real-world CyREST endpoints that are well documented, consume network and customization parameters, and produce actionable results. In this paper, the “Design” section describes the components of Cytoscape Automation and explains their use. The “Implementation” section outlines the details of CyREST construction, and the “Results” section presents concrete examples of Cytoscape Automation benefits. The “Discussion” section compares Cytoscape Automation to other biological workflow environments. Finally, the “Future development” section calls for additional contributions that can expand Cytoscape Automation to improve workflow economics even further. Note that this paper describes CyREST v3.8, which is included with Cytoscape v3.7.0 as a core app, meaning that it is automatically synchronized with Cytoscape by the Cytoscape developer team. Cytoscape Automation requires CyREST v3.8, and we highly recommend that users running Cytoscape versions earlier than v3.7.0 upgrade to v3.7.0. As CyREST evolves, it will be disseminated in new Cytoscape releases and via the Cytoscape App store. CyREST follows semantic versioning guidelines [14], thereby guaranteeing that updates will not break the workflows or automation-enabled apps as it evolves. We highly recommend that independent app developers conform their apps’ evolution to semantic versioning principles, too. DESIGN The leap from the original CyREST implementation to address the broader scope of the Cytoscape Automation initiative required new features and upgraded approaches in a number of technical areas: ❶New CyREST access to Cytoscape apps❷New CyREST access to Cytoscape Command script operations❸Improved documentation infrastructure and content standards❹New interactive CyREST call prototyping❺Consistent mechanisms for calling CyREST and receiving return values❻Improved coverage of core Cytoscape functionality Figure 2 shows the relationship between the Cytoscape Desktop and Cytoscape Automation workflows. The Cytoscape Desktop includes both the Cytoscape core (including CyCommands and CyREST) and apps sourced from the Cytoscape App store. Automation workflows execute outside of Cytoscape but use CyREST to leverage Cytoscape features. Figure 2 is annotated to show the components important in each facet of the Cytoscape Automation design, which are described in this section. Fig. 2Relationship between the Cytoscape Desktop (including CyREST, Cytoscape apps and Cytoscape core) and Cytoscape Automation workflows. Dotted lines indicate command/data flows that pre-date Cytoscape Automation. Solid lines indicate flows created for Cytoscape Automation. New components are in green Note that for a workflow to access Cytoscape functionality, Cytoscape must be running and accessible via HTTP calls from the workflow execution environment. Note that calling a CyREST endpoint requires the use of REST interface functions found in most modern languages. While all CyREST endpoints are accessible in this manner, we have created harmonization libraries for R and Python (described in the “Implementation” section) to enable quick and easy access to common Cytoscape

Read the complete text