Rusteomics 0.2.0 - Building the foundations of a community-driven, open-source data analysis framework in Rust

[di-hardt]

Title

Rusteomics 0.2.0 - Building the foundations of a community-driven, open-source data analysis framework in Rust

Abstract

Most modern programming languages have the concept of libraries or modules, which makes it easy for developers to build frameworks — collections of code designed to solve specific problems. A best practice, especially in research, is to publish the source code in publicly accessible repositories, such as GitHub or GitLab, and distribute it via package registries like PyPI (Python), crates.io (Rust), CRAN (R), or quay.io (Docker). This approach ensures that the code's correctness can be verified and that its functionality is easily reusable.
During the EuBIC developer meeting in 2023, the Rusteomic project was founded with the aim of creating an ultra-fast, reliable, and reusable data analysis framework for (prote-)omics in the modern Rust programming language. The goal is to make this framework accessible to other popular programming languages in research, such as Python, R, and Java. The project began successfully, starting with a brief introduction to Rust, followed by the development of prototypes for a spectrum annotator and a FASTA reader within a few days.
As the number of researchers and developers involved in Rusteomics grew, it became more challenging to align everyone's goals and maintain an overview of the tasks needed before more advanced methods could be implemented, which resulted in a temporary stagnation of development. Before coding can resume, we wish to establish clear guidelines for community engagement, decision-making, and coding standards.
Since Rusteomics is unlikely to be the last project of its kind, the developed guidelines and the process used to define them will be written as general as possible and made available to the broader community for reuse in other projects.

Project Plan

• Guidelines for Community Engagement
  • Feature Requests
    • What should they look like?
    • What information is required?
• Guidelines for Decision Making
  • Feature Discussion
    • Are the necessary requirements already in the framework?
    • Is additional implementation needed before development?
  • Feature Design
    • What will modules, structs, traits, functions, and operators look like?
  • Starting Point
    • How should an agreement on a feature be reached?
    • Does the proposer agree the original goal is intact?
    • Can the community integrate the feature (even if additional work is needed)?
• "Fast Way" Approach: Collect existing tools/methods under this project/organization? Various similar Rust codebases have emerged that could be integrated or harmonized with the Rusteomics project
  • How to contact developers?
  • Should a tool be integrated, and what is required for integration?
• Coding Guidelines
  • How should they be defined?
  • Key considerations (e.g., programming patterns, object-oriented design)
  • Linting/API and Syntax harmonization
  • Language-specific decisions (e.g., async Rust vs. plain Rust)
  • Test the guidelines by discussing and developing the first features.

Technical Details

• Recommended skills: GitHub/GitLab
• Recommended programming languages: Rust
• Affinity to working on large community codebases

Contact information

• Dirk Winkelhardt
  Ruhr-University Bochum, Medical Faculty, Medical Bioinformatics
  dirk.winkelhardt@rub.de
• Ralf Gabriels
  VIB-UGent Center for Medical Biotechnology, VIB, Belgium; Department of Biomolecular Medicine, Ghent University, Belgium