Project Aim
Sequencing experiments generate valuable data that should be shared with the scientific community, encouraging FAIR principles. Yet submission through public repositories remains a major bottleneck. Researchers must navigate a complex multi-step process to decipher schemas, lengthy documentation and manually organise metadata according to strict specifications—all before a single read file reaches the repository. This complexity means that data submission is universally dreaded and avoided depriving the community of potentially valuable datasets.
seqsubmit is an ambitious, community-driven initiative to create an nf-core pipeline to automate submission of diverse genomic data (samples, reads, assemblies etc.) to INSDC databases.
INSDC (International Nucleotide Sequence Database Collaboration) comprises three major repositories which mirror each other’s data:
- ENA (European Nucleotide Archive)
- NCBI GenBank
- DDBJ (DNA Data Bank of Japan)
Each database has its own submission interfaces, tools, and requirements—making truly universal submission tools a significant challenge that requires expertise across multiple platforms and active community involvement.
Goals
For this hackathon, we’ll focus on creating reusable nf-core subworkflows for metagenomic assembly and bin/MAG (Metagenome Assembled Genome) submission to ENA. These subworkflows can be integrated with existing pipelines like nf-core/mag for automated submission.
- Create missing nf-core modules for:
assembly_uploader(MGnify’s tool for metagenomic assembly submission) (Difficulty: Easy/Medium)genome_uploader(MGnify’s tool for MAG/genome submission) (Difficulty: Easy/Medium)webin-cli(ENA’s general submission client) (Difficulty: Easy/Medium)
- Write a
nf-core/magsamplesheet generator fornf-core/seqsubmit(Difficulty: Medium/Hard). - Success: Develop an nf-core subworkflow for uploading metagenomic assemblies with proper metadata handling (Difficulty: Hard).
- Absolute Success: Develop an nf-core subworkflow for uploading MAGs and bins, taking pre-calculated quality statistics (completeness, contamination, coverage) and taxonomic assignments as input (Difficulty: Very Hard).
Let’s make data submission more friendly—one subworkflow at a time! 🚀
Resources
