2025 CMU Biohackathon Report on Genomic and Environmental Effects

34 scientists from nine countries collaborated in March 2025 for a biohackathon exploring machine learning for multimodal computational biology. Their projects, documented in a 7.9 MB PDF, focused on analyzing background genomic and environmental effects on complex traits. The work includes applications of clustering, gene-gene interaction studies, causal discovery in EHR data, and graph-based multiomics analysis.

Use Cases

• Benchmarking machine learning clustering techniques based on the described haplotype similarity analysis.
• Adapting frameworks for gene-gene interaction studies based on the StructLMM application mentioned.
• Developing causal inference pipelines based on the described use of Tetrad software with electronic health records.
• Creating workflow templates for imputation reference panels based on the described nextflow project.
• Designing graph-based analysis tools for multiomics data based on the described pathway intersection diagrams.

Strengths

• Documentation from a collaborative event involving 34 scientists from international institutions.
• Clear focus on six distinct, innovative computational biology projects as described.
• Released under a permissive CC-BY-4.0 license for reuse.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• The 7.9 MB size suggests the PDF is a report, not a primary dataset, limiting direct analysis.

Provenance

Source

Hackathon hosted by DNAnexus and Carnegie Mellon University Libraries.

Collection Method

Collaborative projects developed during the March 2025 event.

Time Range

March 2025.

Freshness

Last updated 2026-05-07 18:37:53; freshness should be verified.

Geography

Participants from the United States, Ireland, United Kingdom, Switzerland, France, Germany, Spain, India, and Australia.