2024 SPARC FAIR Codeathon

The codeathon showcase features the participating teams presentations & shows their experiences participating in a global virtual codeathon.

Teams shared their experiences developing their innovative projects to make SPARC data, tools, & resources more Findable, Accessible, Interoperable, & Reusable during the 3-day in-the-cloud event.

Seven Projects:

SPARC.RL

Grand Prize, $15,000 USD

SPARC.RL Offers a proof of concept toolchain that simplifies the training of state-of-the-art reinforcement learning agents, even for those without deep expertise in the field.

oSPARC-Hub

Grand Prize, $15,000 USD

The oSPARC Hub framework is a comprehensive suite of SPARC pipelines, modules, and applications designed to enhance the capabilities of the oSPARC platform.

sparc-spy

2nd Prize, $7,000 USD

A cross-platform Python visualisation tool called the SPARC Scaffolding in Python (SPARC-SPy) to run within o2S2PARC that can produce VTK visualisations from data scaffolds.

SPARCTA

3rd Prize, $3,000 USD

A web-based viewer for TIFF files with annotation capabilities, called SPARCTA.

sparc-assemble

A python tool to automatically assemble models and tools into workflows, for processing and interpreting SPARC datasets in accordance with FAIR principles.

DataFusion

A service to enhance the usability of SPARC by converting datasets into neurodata without borders and facilitating hypothesis-driven studies by providing preliminary analyses.

sPARcRNA_Viz

Single Cell RNA-sequencing differential expression analysis and interactive visualization service on oSparc.

Selected Project Ideas

• Extensible tools for the extraction of information from datasets (e.g. csv spreadsheets, json data) and population of a quantitative data base.

• Development of connectivity visualization/interaction tools (e.g. using cytoscape widgets in Jupyter; D3 in Javascript).

• Derive computational models from the vagus scaffold to explore electrical conductivity using oSPARC S4Llite.

• Create a tiff image viewer component for viewing .tiff files on the portal.

• Simple Protocols.io widget that provides a quick overview of the protocol that could be integrated with the portal.

• Analysis of SPARC gene expression data utilizing R, octave or matlab in oSPARC.

• Encoding experimental data into a Scaffold VTK export for visualisation/simulation on oSPARC.

• Leverage the existing algolia and discover indices to Implement a more refined and/or visual user-friendly query interface for finding and navigating datasets and/or dataset files.

• Use the SPARC Python Client to perform complex queries on the SPARC Portal (e.g. finding datasets containing certain files types) and reuse that data.

• Turn your code/model into a module for the oSPARC platform, by using the cookiecutter-osparc-service library.

FAIR Codeathon FAQs:

Who can participate? Teams will greatly benefit from people who possess any of the following skills:

• Data mining, image and text analysis
• Working knowledge of scripting languages (e.g., Shell, Python, R)
• Familiarity with methods for manipulating and/or analyzing large datasets (AI/ML, computational modeling, etc.)
• Developing bioinformatics code, pipelines or tools
• Data visualization
• Knowledge graphs
• Web development
• Understanding of neuroscience and/or neurostimulation

Do I have to lead a team? You can choose to lead your project team, recommend someone, or we can try to find a suitable team lead. Providing a designated team lead dramatically increases the probability that we will select the project for the codeathon.

Click here to join a team: [Click Here]

Do I need to assemble a team? No. We will create working groups of five to six individuals who have various backgrounds and relevant expertise to work on each project.

What are my responsibilities as a team lead? The team leader will coordinate a group of 5-6 people in defining the project and producing clear vision for developing a solution. To accomplish this goal, the team lead must define and delegate tasks, incorporate team members’ ideas to accomplish the goal, and ensure the team’s success.

What if I only want to participate? Applications for those who would like to participate in the codeathon will be available in late June. Here is the link to sign up.

What will a typical day be like? We will meet regularly as a group throughout the codeathon, exact timing to be determined as teams are formed. SPARC Data & Resource Center teams will host regular office hours enabling teams to seek guidance.

What will we build? We will make all pipelines, other scripts, software, and programs generated in this codeathon available on a dedicated public GitHub organisation.

Teams may submit manuscripts describing the design and use of the software tools they created to an appropriate journal such as the F1000Research hackathons channel, GigaScience, or PLoS Computational Biology.

What is FAIR? The volume of publicly available data continues to rise exponentially, but the capacity for fully employing this data is being hampered by a series of limitations. FAIR is a very powerful initiative that has taken root worldwide. The initiative has the potential to significantly increase the value of life science data sets. While the concept shares some commonality with the semantic web, FAIR data goes further to expand opportunities for knowledge-sharing and value. Here are four foundation papers on this exploding field:

• The FAIR Guiding Principles for Scientific Data Management and Stewardship
• The Healthcare and Life Sciences Community Profile for Dataset Descriptions
• Publishing FAIR Data: An Exemplar Methodology Utilizing PHI-Base
• 5 Principles for FAIR Software

More information about SPARC:

There are several tools and resources available that may be leveraged when designing the projects. Codeathon projects should result in code, tools (see SODA for an example), datasets, or other outputs which are open and freely available.

Investigators are publishing lots of FAIR data from a range of species, and spanning all the major visceral organs and peripheral nerves, as they seek to better understand the autonomic nervous system. The data being collected includes microscopy, electrophysiological and mechanical time series, single-cell RNASeq, functional MRI, and more.

SPARC data is highly curated, ensuring the data is published in a FAIR manner and backed by a semantic knowledge base. SPARC data is further enriched by being mapped to 2D “flatmaps” that enable visual exploration of the topological anatomy of the peripheral nervous system. Where possible, data is also mapped to 3D organ scaffolds to provide a common coordinate system enabling comparison across subjects, species, and protocols as well as interactive environments to aid understanding and interpretation of SPARC data. The open-source tools for mapping data to scaffolds are available.

One special data resource, called “simulations”, consists of computational models and data analysis pipelines. These simulations can be run on o²S²PARC, which was designed to host, modularize, and ensure the reproducibility of simulations contributed by researchers. This is achieved by archiving contributed simulation code along with the code’s execution environment with versioning.

SPARC data, simulations, and maps are published on the SPARC Portal, an open-source platform for finding, exploring, visualizing, interacting with, and accessing SPARC data and associated computational models and analyses.

Legal
Participants retain ownership of all intellectual property rights (including moral rights) to the code submitted to as well as developed in the codeathon. Employees of the U.S. Government attending as part of their official duties retain no copyright to their work and their work is in the public domain in the U.S. The Government disclaims any rights to the code submitted or developed in the codeathon. Participants agree to publish the code and any related data on GitHub.

Please feel free to contact the FAIR Codeathon team if you have questions or need more information: fair-codeathon-support@sparc.science.