NCBI's Virus Discovery Hackathon: Engaging Research Communities to Identify Cloud Infrastructure Requirements

Area of Science:

• Virology
• Bioinformatics
• Computational Biology
• Genomics

Background:

• Vast amounts of viral data are underutilized in public metagenomic datasets.
• Extracting this data could create a valuable index for the virological research community.
• The feasibility of large-scale, complex biological data analysis in a hackathon setting was unexplored.

Purpose of the Study:

• To test the hypothesis that a hackathon could generate crowd-sourced analysis and processing pipelines for complex viral metagenomic data.
• To extract and analyze viral data from the National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA).
• To identify bottlenecks and solutions for analyzing large biological datasets using cloud infrastructure.

Main Methods:

• A three-day hackathon involving over 40 participants from six countries.
• Pre-assembled metagenomic data from NCBI SRA was filtered and processed.
• Contigs were aligned against known viral genomes using BLAST, phylogenetically clustered, and assigned metadata.
• Subsets of contigs were screened for viral genes and domains.

Main Results:

• Over 4.2 million (Mio) contigs were generated from 2953 SRA datasets.
• 360,000 contigs were assigned viral domains, and 4400 were screened for viral genes.
• Conservative assemblies improved analysis, wrapper scripts enhanced software performance, and cloud infrastructure facilitated collaboration.

Conclusions:

• Hackathons are effective for crowd-sourcing complex biological data analysis pipelines.
• Cloud infrastructure is crucial for enabling effective collaboration among diverse researchers.
• The study identified key insights into SRA data analysis and optimized bioinformatic workflows for large-scale virological research.