How israamanaziz/schizophrenia-exosomal-mirna-biomarker-pipeline works
This repository documents a research workflow for prioritizing plasma exosomal miRNA biomarkers in first-episode schizophrenia. The principal path runs from public sequencing data through preprocessing and quantification, count-matrix construction, differential-expression analysis, machine-learning prioritization, target and network analysis, and enrichment. The README documents these stages, but no source excerpts were provided; unsampled stages therefore use only documented links and have no asserted implementation path.
Main entry points: 04_functional_enrichment.Rmd GO and KEGG enrichment · 02mirna_deseq2_analysis.Rmd Differential expression · 01miRNA_count_matrix_construction.ipynb Count matrix construction · miRNA_countmatrix.csv miRNA count matrix
Components 17
Expression Analysis
- Count matrix construction
01miRNA_count_matrix_construction.ipynb - miRNA count matrix
data/miRNA_countmatrix.csv - Sample metadata
data/sample_info.csv - Differential expression
02mirna_deseq2_analysis.Rmd
Biomarker Prioritization
- Feature selection
03miRNA_ML _analysis.Rmd - Candidate biomarkers
Functional Interpretation
- miRDB prediction
- Target genes
- Cytoscape network
- Hub genes
- GO and KEGG enrichment
04_functional_enrichment.Rmd - Biological interpretation
Other components
- GEO datasets
- Galaxy
- Fastp
- miRDeep2
- clusterProfiler
Connections 16
- GEO datasets → Galaxy: provides datasets
README.md - Galaxy → Fastp: runs preprocessing
README.md - Galaxy → miRDeep2: runs quantification
README.md - miRDeep2 → Count matrix construction: provides quantifications
README.md - Count matrix construction → miRNA count matrix: constructs matrix
README.md - miRNA count matrix → Differential expression: supplies counts
README.md - Sample metadata → Differential expression: supplies metadata
- Differential expression → Feature selection: provides results
README.md - Feature selection → Candidate biomarkers: prioritizes candidates
README.md - Candidate biomarkers → miRDB prediction: submits candidates
README.md - miRDB prediction → Target genes: predicts targets
README.md - Target genes → Cytoscape network: forms interaction network
README.md - Cytoscape network → Hub genes: identifies hubs
README.md - Hub genes → GO and KEGG enrichment: supplies genes
README.md - GO and KEGG enrichment → clusterProfiler: uses
README.md - GO and KEGG enrichment → Biological interpretation: supports interpretation
README.md