Train genomic region embeddings and analyze single-cell ATAC-seq data with Python.
npx skills add https://github.com/hxk622/TokenDance --skill geniml-hxk622
Please help me install this Agent Skill. Skill: geniml Source: https://github.com/hxk622/TokenDance/tree/main/backend/app/skills/builtin/scientific/bioinformatics/geniml Command: npx skills add https://github.com/hxk622/TokenDance --skill geniml-hxk622
This Skill simplifies complex machine learning tasks on genomic interval data (like BED files) and single-cell ATAC-seq data, making advanced bioinformatics analysis accessible.
Use the geniml skill to train Region2Vec embeddings on BED files in the 'bed_files/' directory using 'universe.bed' as a reference.
High-intent search queries and answers about installing and using this skill.
Train region embeddings on BED files using the Region2Vec or BEDspace models. These models learn unsupervised representations of genomic regions and their metadata, enabling dimensionality reduction, similarity searches, and clustering for downstream genomic interval analysis.
Cluster cells from scATAC-seq data by generating cell embeddings with the scEmbed model. This approach supports unsupervised learning for dimensionality reduction and cell-type annotation based on chromatin accessibility patterns.
Build a consensus peak set from multiple ChIP-seq BED files to identify common regulatory regions. This process constructs a robust reference peak set across multiple experiments, serving as a reliable universe for downstream embedding training.
Predict cell types based on chromatin accessibility patterns by training machine learning models on single-cell ATAC-seq data. The toolkit generates cell embeddings that capture accessibility features, enabling clustering and cell-type annotation.
Training Region2Vec embeddings requires a reference BED file, such as a universe file. This reference defines the genomic scope, while the model learns representations from the intervals and metadata within the provided BED file directory.
Region2Vec applies unsupervised representation learning specifically to genomic interval data rather than sequences. It captures the relationships between genomic regions and associated metadata, enabling similarity searches and clustering that interval-specific approaches facilitate.
shushuznBuild machine learning embeddings from genomic interval BED files.
tassiovaleStandardize single-cell RNA-seq workflows with Scanpy for QC, clustering, and annotation.
RamanEbrahimiProcess genomic interval BED files for machine learning tasks.
Lord1EgyptTrain machine learning models on genomic interval data from BED files.
