skills.rest
JosephWoodall
JosephWoodallCOMMUNITY
Explore all skills by this author →
𝕏@josephwoodall_GitHub

geniml

Build unsupervised genomic interval embeddings from BED files with Region2Vec, BEDspace, scEmbed, and universe-building workflows.

1|Updated Mar 19, 2026
Search Tags:#embeddings#genomics#tokenization#unsupervised-learning#scatac-seq#genomic-intervals#bed-files
One-click install
npx skills add https://github.com/JosephWoodall/noosphere --skill geniml-josephwoodall
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: geniml
Source: https://github.com/JosephWoodall/noosphere/tree/main/.agent/skills/geniml
Command: npx skills add https://github.com/JosephWoodall/noosphere --skill geniml-josephwoodall

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Geniml provides a Python-based framework for building unsupervised embeddings from genomic interval data (BED files), enabling similarity searches, clustering, and downstream ML tasks on genomic regions, single-cell data, and metadata.

Core Features & Use Cases

Geniml offers Region2Vec for learning embeddings of genomic regions from BED collections, BEDspace for joint region-and-metadata embeddings, scEmbed for single-cell ATAC-seq embeddings, and Universe-building (consensus peaks) for standardized tokenization references; it also includes tokenization, evaluation, caching, and search utilities to support scalable genomics ML workflows.

Quick Start

Train a Region2Vec model on BED files to generate genomic region embeddings for downstream ML tasks.

Frequently Asked Questions about geniml

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

FAQPage Schema
How do I generate embeddings for genomic intervals from BED files?▼

To generate genomic interval embeddings from BED files, you train a Region2Vec model on bulk BED collections. This unsupervised approach produces region representations that enable similarity searches, clustering, and downstream ML analyses.

Can I build embeddings for single-cell ATAC-seq workflows?▼

Yes, you can build single-cell ATAC-seq embeddings using the scEmbed workflow. It generates unsupervised representations from single-cell genomic interval data to support metadata-aware similarity searches and clustering.

What is a universe-building pipeline for genomic tokenization?▼

A universe-building pipeline creates consensus peaks from BED files to serve as a standardized tokenization reference. This process allows you to consistently map genomic intervals into tokens for scalable ML workflows.

Does this approach support joint embeddings for genomic regions and metadata?▼

Yes, the BEDspace workflow builds joint region-and-metadata embeddings. This approach integrates genomic intervals with associated metadata into a unified space for comprehensive downstream ML analyses.

Do I need a Python environment to tokenize BED files for machine learning?▼

Yes, you need a Python environment to tokenize BED files against a universe and run the embedding workflows. The framework relies on Python to support Region2Vec, scEmbed, and BEDspace processing pipelines.

Related Skills

Explore More Data & Analytics Skills→
rubenslivrubensliv
15

geniml

Generate unsupervised embeddings from genomic interval BED files for ML analyses.

Community
Advanced
AlterLab-IEUAlterLab-IEU
58

alterlab-geniml

Build unsupervised embeddings from genomic interval data with Geniml.

Community
Advanced
OwnLabAIOwnLabAI
21

geniml

Generate unsupervised embeddings for genomic intervals from BED files.

Official
Intermediate
fuzzy-dynamicsfuzzy-dynamics
1

geniml

Trains unsupervised embeddings for genomic intervals and metadata from BED files.

Official
Advanced