Kiana-Asgari/MultiOmics_Alterations_HUUC

Multi-omics statistical analysis pipeline for HIV-exposed mother-newborn cohorts (Kenya, Zambia). # MultiOmics Alterations in HUUC Statistical analysis of mother–newborn lipidome and metabolome cohorts (Kenya and Zambia), comparing HIV‑exposed and HIV‑unexposed pregnancies and newborns. The pipeline runs a regularized logistic-regression model with stability selection to identify the biomarkers most consistently associated with HIV exposure, and renders the corresponding box plots, stability/lasso paths, and volcano plots from cached results. ## Top-level directories | Path | Purpose | | --- | --- | | `configs/` | YAML-driven configuration (cohorts, file mappings, marker name maps). | | `common/` | Shared library: data loading, feature engineering, plotting / modeling utilities. | | `maternal_HIV_effects/` | Maternal-side analysis (load, describe, stability-select, lasso-path, volcano). | | `neonatal_HIV_exposure_biomarkers/` | Neonatal-side analysis (HEEL-prick data, stability path, volcano, violin). | | `results/` | Generated PCA / PLSR / ROC / volcano / UMAP figures. | | `script.py` | CLI entry point that re-renders figures from cached selection results. | | `utils.py` | Verbose-printing helper used by the data loaders. | | `requirements.txt` | Python dependencies. |