How does Geniml handle consensus peak sets?

Geniml includes a 'Universe Building' capability that uses statistical methods like Maximum Likelihood (ML) and Hidden Markov Models (HMM) to create reference peak sets from BED collections.

What is Geniml used for?

Geniml is used for building machine learning models on genomic interval data, specifically BED files, to perform tasks like region embedding, single-cell analysis, and metadata-aware searches.

What are the core utilities provided by Geniml?

Geniml provides utilities for BED file caching (BBClient), genomic randomization (BEDshift), evaluation metrics for embeddings, and neural search backends (Text2BedNN).

Can Claude help with single-cell ATAC-seq analysis using Geniml?

Yes, this skill enables Claude to guide users through the scEmbed workflow, including tokenizing cells, training models, and integrating results with scanpy for clustering.

What is Region2Vec in the context of this skill?

Region2Vec is a method within Geniml that uses word2vec-style learning to create unsupervised embeddings of genomic regions for similarity analysis and feature extraction.

Geniml Genomic ML

Name: Geniml Genomic ML
Author: BbgnsurfTech

byBbgnsurfTech

•

数据科学与机器学习

Applies machine learning to genomic interval data for training region embeddings, analyzing scATAC-seq, and building consensus peak universes.

Geniml is a specialized Python toolkit designed for applying unsupervised machine learning models to genomic interval data, primarily BED files. It enables researchers and bioinformaticians to generate high-quality embeddings for genomic regions using Region2Vec, integrate metadata with region sets via BEDspace, and perform sophisticated single-cell chromatin accessibility analysis with scEmbed. By automating the tokenization of genomic regions and providing statistical methods for building reference 'universes,' Geniml streamlines the transition from raw genomic coordinates to actionable machine learning features for clustering, similarity searches, and downstream biological analysis.

主要功能

01scEmbed for cell-level embeddings and clustering of single-cell ATAC-seq data

02Comprehensive utilities for BED file caching, randomization, and embedding evaluation

03Statistically rigorous consensus peak (universe) building using CC, ML, and HMM methods

048 GitHub stars

05Region2Vec for unsupervised genomic region embeddings and similarity analysis

06BEDspace for joint embedding of genomic regions and experimental metadata

使用场景

01Cell-type annotation and dimensionality reduction for scATAC-seq workflows

02Feature engineering for downstream supervised learning tasks in computational biology

03Building searchable genomic databases using metadata-aware embeddings

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add bbgnsurftech/claude-skills-collection geniml

For use in Claude.ai and ChatGPT

Download Skill