geniml

Train machine learning models on genomic interval BED files with geniml.

13|3|Updated Jun 10, 2026

Search Tags:#machine learning #dimensionality reduction #single-cell ATAC-seq #genomic interval data #genomic embeddings

Ready to run

npx skills add https://github.com/tassiovale/claude-code-kit --skill geniml-tassiovale

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: geniml
Source: https://github.com/tassiovale/claude-code-kit/tree/main/skills/geniml
Command: npx skills add https://github.com/tassiovale/claude-code-kit --skill geniml-tassiovale

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires geniml, PyTorch, scikit-learn, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This skill addresses the challenge of applying machine learning to genomic interval data (BED files), enabling users to perform various genomic analyses and tasks efficiently.

Core Features & Use Cases

Genomic Embeddings: Learn embeddings for genomic regions, single cells, and metadata labels.
Dimensionality Reduction: Reduce the dimensionality of genomic data for easier analysis.
Single-Cell Analysis: Analyze single-cell ATAC-seq data for clustering, annotation, and integration with scanpy.
Universe Building: Build consensus peak sets from BED file collections for tokenization and analysis.
Use Case: For a researcher looking to analyze scATAC-seq data, this skill provides tools like scEmbed to generate cell embeddings and integrate with popular analysis frameworks like scanpy.

Quick Start

Train an embedding model for scATAC-seq data using the scEmbed tool from the geniml skill.

Frequently Asked Questions about geniml

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I apply machine learning to BED files for genomic interval analysis?▼

To apply machine learning to BED files for genomic interval analysis, you can use tools that build consensus peak sets from BED collections, tokenize the intervals, and train models to generate embeddings for downstream analysis.

Can I generate cell embeddings from single-cell ATAC-seq data?▼

Yes, you can generate cell embeddings from single-cell ATAC-seq data using the scEmbed tool, which trains embedding models on genomic intervals and integrates the resulting cell representations with the scanpy analysis framework.

What Python dependencies are required for genomic machine learning workflows?▼

Python dependencies required for genomic machine learning workflows include geniml for the core tools, PyTorch for model training, and scikit-learn for dimensionality reduction and analysis tasks.

How do I reduce dimensionality in genomic interval datasets?▼

You can reduce dimensionality in genomic interval datasets by training embedding models on consensus peak sets to learn lower-dimensional representations of genomic regions and single cells for easier downstream analysis.

Does geniml integrate with scanpy for single-cell clustering?▼

Yes, geniml integrates with scanpy by generating cell embeddings from single-cell ATAC-seq data, which can then be used directly within the scanpy framework for clustering, annotation, and further analysis.

What is a universe in genomic interval machine learning?▼

A universe in genomic interval machine learning is a consensus peak set built from a collection of BED files, which serves as the vocabulary for tokenizing genomic regions before training embedding models.