Bio Structure Annotation

When to use

• Structure prediction and structure-based annotation.

Prerequisites

• Tools installed via pixi (see pixi.toml).
• Reference DB root: /media/shared-expansion/db/ (wsu; override per machine branch).
• Protein FASTA inputs are available.

Inputs

• proteins.faa (FASTA protein sequences)

Outputs

• results/bio-structure-annotation/structures/
• results/bio-structure-annotation/structure_hits.tsv
• results/bio-structure-annotation/structure_report.md
• results/bio-structure-annotation/logs/

Steps

1. Run fast embedding screen (tm-vec).
2. Predict structures (boltz or colabfold) as needed.
3. Search structures with Foldseek and annotate hits.

QC gates

• Prediction success rate meets project thresholds.
• Search hit thresholds meet project thresholds.
• On failure: retry with alternative parameters; if still failing, record in report and exit non-zero.

Validation

• Verify proteins.faa is non-empty and amino acid encoded.
• Verify Foldseek databases exist under the reference root.

Tools

• tm-vec v1.0.3
• boltz v2.2.1
• colabfold v1.5.5
• foldseek v10-941cd33

Paper summaries (2023-2025)

• summaries/ (include example use cases and tool settings used)

Tool documentation

• TM-Vec - Fast protein structure embedding and similarity search
• Boltz - AI-based protein structure prediction
• ColabFold - Fast protein structure prediction using AlphaFold2
• Foldseek - Fast structure-based protein search

References

• See ../bio-skills-references.md