omics-ml-classifier

What problem does it solve?

Train supervised machine learning classifiers on omics data to predict phenotypes and identify biomarkers with rigorous evaluation and interpretable outputs.

Core Features & Use Cases

• Supports multiple models: random_forest, logistic, svm, and xgboost, with cross-validated performance reporting.
• Works with two-class and multi-class labels and provides feature importances and ROC curves.
• Use cases include classifying cancer subtypes from expression data, predicting treatment response from multi-omics features, and ranking biomarkers via feature importance.
• End-to-end workflow: accepts a numeric feature matrix and label file, preprocesses, trains, evaluates, and outputs metrics and plots.

Quick Start

Train a classifier on your omics feature matrix and evaluate its performance using cross-validated metrics and a held-out test set.