How collaborativebioinformatics/clinical_svs works
Project type and purpose
• This is a hybrid “full-stack” bioinformatics pipeline + interactive web app.
• Purpose: ingest VCFs of structural variants (SVs), annotate them with clinical/genomic information, perform enrichment analyses, and present results either via command-line outputs (annotated VCF/CSV + QC graphs) or through a Shiny web interface.
File structure
Top-level directories:
• R/ ← R-based annotation modules & CLI scripts
• VCF_parser/ ← Python script to preprocess VCFs
• shiny/ ← Shiny app (app.R, assets, R modules)
• demo_outputs/ ← example enrichment outputs
• testdata/ ← sample inputs and expected outputs
Key scripts:
• R/prepare_annotation_data.R ← builds annotation_data.RData (data prep)
• R/annotate_vcf.R ← master CLI annotator
• R/annotate_*.R ← modular annotation functions (genes, frequency, clinical)
• R/geneFunctionalAnnotation.R ← enrichment (KEGG, Reactome, DO)
• shiny/app.R + shiny/*.R ← UI + server sourcing same modules
• VCF_parser/VCFanalyse.py ← Python-based VCF filtering/parsing
Config/build: no Docker files, but Rscript wrappers and Python script. Dependencies declared in README.
README for additional insights
• Two usage modes: (1) CLI workflow (SV calling → annotation → QC graphs) (2) Interactive Shiny app
• Lists external data sources: gnomAD-SV catalog, dbVar clinical SVs, OMIM, Gencode v35, biomedical ontologies