Facilitating bioinformatics reproducibility with QIIME 2 Provenance Replay

Area of Science:

• Bioinformatics
• Computational Biology
• Scientific Reproducibility

Background:

• Scientific research reproducibility is crucial but challenging in bioinformatics.
• Complex workflows and large datasets hinder reproducibility.
• Biologists often lack training in documenting bioinformatics analysis steps, leading to missing information.

Purpose of the Study:

• To address the challenge of bioinformatics reproducibility.
• To present a software tool that automates provenance tracking.
• To guide bioinformatics developers in implementing similar reproducibility features.

Main Methods:

• Developed Provenance Replay, a software tool.
• Implemented automated provenance tracking within the QIIME 2 bioinformatics platform.
• Generated executable code from QIIME 2 analysis results.

Main Results:

• Provenance Replay successfully generates executable code from QIIME 2 results.
• The tool automates the recording of bioinformatics analysis steps.
• Barriers to bioinformatics reproducibility are significantly reduced.

Conclusions:

• Automated provenance tracking is key to enhancing bioinformatics reproducibility.
• Provenance Replay offers a practical solution for QIIME 2 users.
• The study provides insights for developing future reproducibility tools in bioinformatics.