The Galaxy platform for accessible, reproducible, and collaborative data analyses: 2024 update.

Galaxy Community. · 2024

Read The Galaxy platform for accessible, reproducible, and collaborative data analyses: 2024 update. by Galaxy Community. (2024) 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-pmc11223835.txt.

Introduction

Galaxy Community reports the 2024 state of a global platform for data analysis in biology in this paper. Galaxy lets researchers run complex analyses through a web interface without local computing power. Most services are free to use, and the user base broadens each year. Stability, tool variety, training, and support draw users in.

Recent work applies user experience design to improve accessibility and tool discovery. A redesigned ToolShed, the platform's app store, is part of that push. Two technical directions stand out: access to graphics processors for cutting-edge methods, and support for licensed tools. The Galaxy Training Network has grown its course portfolio and linked lessons directly to tools. Job scheduling and the interface were improved. The authors also describe resourcing public services to run more shared workflows.

Catalog record

CREATORGalaxy Community.
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE53 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC11223835

Browse the full library · About the Tower of Babel

Full text

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

The Galaxy platform for accessible, reproducible, and collaborative data analyses: 2024 update Published in: Nucleic Acids Research DOI: 10.1093/nar/gkae410 ABSTRACT Abstract Galaxy (https://galaxyproject.org) is deployed globally, predominantly through free-to-use services, supporting user-driven research that broadens in scope each year. Users are attracted to public Galaxy services by platform stability, tool and reference dataset diversity, training, support and integration, which enables complex, reproducible, shareable data analysis. Applying the principles of user experience design (UXD), has driven improvements in accessibility, tool discoverability through Galaxy Labs/subdomains, and a redesigned Galaxy ToolShed. Galaxy tool capabilities are progressing in two strategic directions: integrating general purpose graphical processing units (GPGPU) access for cutting-edge methods, and licensed tool support. Engagement with global research consortia is being increased by developing more workflows in Galaxy and by resourcing the public Galaxy services to run them. The Galaxy Training Network (GTN) portfolio has grown in both size, and accessibility, through learning paths and direct integration with Galaxy tools that feature in training courses. Code development continues in line with the Galaxy Project roadmap, with improvements to job scheduling and the user interface. Environmental impact assessment is also helping engage users and developers, reminding them of their role in sustainability, by displaying estimated CO2 emissions generated by each Galaxy job. INTRODUCTION User demand for an easily accessible data analytics service, deployed on computing infrastructure capable of meeting the needs of complex computing in research, has resulted in the Galaxy Project supporting, in its 19th year of ongoing operation, a rapid increase in throughput globally (1,2). Galaxy provides analytical tools that can be used individually or linked into complex workflows with intermediate data outputs capable of triggering logic conditionals within the workflow. Recent enhancements allow researchers to run workflows on data of variable quality, and have the workflow buffered to systematically explore experimental variability (https://gxy.io/GTN:T00164). Large scale research is necessarily collaborative, and Galaxy's capacity to both securely share and publish data and workflows supports efficient collaboration, training, and data reuse. Recent changes to the Galaxy user interface discussed below have made sharing more visible. Collectively, the usegalaxy.* services in the United States, Australia, and Europe have amassed >500 000 registered users, and supports >11 000 individual users running >1 000 000 jobs on average each month of 2023. Usegalaxy.* service statistics are publicly available at https://status.galaxyproject.org/, with detailed operational data for Australia and Europe at https://stats.usegalaxy.org.au/ and https://stats.galaxyproject.eu/, respectively. Users have access to >9000 scientific tools, supporting >400 different types of input data, enabling a wide variety of analyses in both the life and physical sciences, including astronomy, genomics, proteomics, metabolomics, materials science, imaging, and cytometry. Efficient, reproducible complex analytical pipelines can be created by joining tools from any domain with ‘noodles’ on the workflow canvas. Analysis outputs can be explored with >50 types of inbuilt visualizations, and a simple URL can be shared with collaborators, which encapsulates all the data, analyses settings, tool versions, and workflows needed for replication. RESEARCH DRIVEN SOLUTIONS Galaxy makes thousands of third-party open-source analysis packages easy to use, and interoperable without any user supplied code. For any new analysis package to become a tool, a developer prepares a Galaxy wrapper once, and uploads it to the sharable Galaxy global tool ‘appstore’ called the Galaxy Toolshed (https://toolshed.g2.bx.psu.edu/). Each Galaxy service supports a core common tool set, and offers a wide range of other tools, the exact combination of which is driven by user demand. The tools are then categorized by scientific use and/or datatypes involved. Further, the option to host tools on Galaxy is conditional on the tools’ computational needs (e.g. GPGPUs, high-memory), licence stipulations of tool use, and the ability for it to be wrapped as a standard or interactive tool, all of which have been improved in the latest Galaxy updates (https://docs.galaxyproject.org/en/master/releases/index.html). Galaxy Labs/sites/subdomains The Galaxy Toolshed now hosts over 9500 distinct and modular software packages, i.e. tools, available to Galaxy Administrators for easy installation on any Galaxy service. The breadth of analytical options available as installed tools can add value for end users but can also be overwhelming. User feedback, both active formal UX documentation and passive user-initiated feedback, have identified that even a fraction of the tools hosted on a Galaxy service can confuse the process of finding any specific tool. For example, a user looking for a singular tool on the usegalaxy.* servers have to navigate through sets of 1770 (Galaxy US), 3320 (Galaxy EU) and 1730 (Galaxy AU) tools. This can be daunting, even when considering the help provided through tool categorisation and EDAM ontology labels (3). Galaxy Europe first provided a solution to empower researchers with a common interest or who undertake a set of activities frequently. The subdomains focus on a particular research domain or technology modality (Figure 1). The content, tools and resources available are ‘tailored’ to each domain - i.e. making sure that the resources are a good fit for routine real-world research practice. Galaxy Australia also now makes use of this option, naming its offering Galaxy Labs (Figure 1). These labs offer a concentration of tools, workflows, and resources allowing new and regular practitioners of that field ready access to the most common options they need, whilst still offering all the other Galaxy features they could use. Importantly, a user logged in on a Galaxy offering labs/subdomains has full access to their data (histories), workflows, shared data, across all labs and the main service page. Galaxy Labs also align with Galaxy Project's strategic initiative to support global research consortia, such as the Vertebrate Genome Project (VGP) (4) and Earth BioGenome Project (EBP) (5). Important regional examples include the adoption of Galaxy within the European Partnership for the Assessment of Risks from Chemicals (EU-PARC; https://www.eu-parc.eu/) (6), WP4 Task 4.3.1.d, as a platform of choice for processing small molecule mass spectrometry datasets, and separately for processing of mass spectrometry datasets generated via the Czech node (coordinator) of the European Environmental Exposure Assessment Research Infrastructure (EIRENE-CZ; https://www.eirene-ri.eu/). In Australia, the Threatened Species Initiative uses Galaxy Australia as its primary genomics analytical service (https://threatenedspeciesinitiative.com/genome-assembly/). [Figure: Examples of Galaxy Labs/subdomains. Researchers can quickly access a concentration of domain-specific tools, workflows, support, and training through Galaxy Labs or Galaxy subdomains. Top: the Genome Lab and Proteomics Lab on Galaxy Australia, https://genome.usegalaxy.org.au and https://proteomics.usegalaxy.org.au. Bottom: the Single Cell Omics subdomain on Galaxy Europe, https://singlecell.usegalaxy.eu/ and https://hicexplorer.usegalaxy.eu.] GPGPU-supported tools Increased tool complexity and capacity is evident in Galaxy through recent tool offerings utilising GPGPU infrastructure. Tool execution environment resourcing is not a consideration for Galaxy users but is easily configured by Galaxy infrastructure providers. GPGPU-based tools (AlphaFold2.0/multimer (7), ChatGPT (8) and AI-based tools run in JupyterLab (7)) are highly requested and are delivered through local access to GPGPU or through commercial cloud provision, across the usegalaxy.* services. In the case of Galaxy Australia, this work necessitated the deployment of a Pulsar on the Australian instance of Microsoft Azure. The demand for AlphaFold2.0/multimer in Australia has branched into utilization of NVIDIA GPUs and more recently the code has been adapted to also work on AMD GPUs, to decouple the tool use from a specific set of hardware requirements and to allow the tool to be deployed on a greater range of worker node configurations. Within

Read the complete text