This skill should be used when working with genomic interval data (BED files) for machine learning tasks. Use for training region embeddings (Region2Vec, BEDspace), single-cell ATAC-seq analysis (scEmbed), building consensus peaks (universes), or any ML-based analysis of genomic regions. Applies to BED file collections, scATAC-seq data, chromatin accessibility datasets, and region-based genomic feature learning.
Geniml is a specialized Python package designed to facilitate machine learning tasks involving genomic interval data, particularly in the form of BED files. It addresses the challenges associated with analyzing complex genomic datasets and deriving meaningful insights through advanced machine learning techniques. By utilizing unsupervised learning methods, Geniml enables researchers to extract embeddings from genomic regions, thereby allowing for more effective similarity searches, clustering, and other downstream analyses that are crucial in genomic studies.
The core capabilities of Geniml include training genomic region embeddings through Region2Vec, generating joint embeddings for genomic regions and metadata with BEDspace, and creating single-cell chromatin accessibility embeddings using scEmbed. Additionally, it offers tools for building consensus peak sets from BED file collections and a range of utilities for caching, randomization, evaluation, and tokenization. These features empower users to conduct comprehensive analyses while maintaining the integrity and context of the genomic data.
Geniml is particularly valuable for bioinformaticians, geneticists, and data scientists working with genomic data who require sophisticated tools for machine learning applications. Its versatile capabilities make it suitable for tasks such as training region embeddings, analyzing single-cell ATAC-seq data, and constructing reference peak sets, thus supporting a wide range of research objectives in genomics and epigenomics.
You can install Geniml using pip with the command: `pip install geniml`. For additional ML dependencies, use `pip install 'geniml[ml]'`. For the development version, use `pip install git+https://github.com/databio/geniml.git`.
Geniml is designed to work with BED files, single-cell ATAC-seq data, and chromatin accessibility datasets.
Yes, Geniml integrates well with popular machine learning frameworks, particularly PyTorch, which is used for various ML tasks.
Geniml requires Python and should be run in an environment where necessary dependencies, such as PyTorch, are installed.
While Geniml offers extensive capabilities for genomic data analysis, users should be aware that the effectiveness of the models may depend on the quality and size of the input data.
Quick Setup:
.claude/skills/Repository
K-Dense-AI/claude-scientific-skills