Build machine learning embeddings from genomic interval BED files.
npx skills add https://github.com/shushuzn/Rairos --skill geniml-shushuzn
Please help me install this Agent Skill. Skill: geniml Source: https://github.com/shushuzn/Rairos/tree/main/skills/geniml Command: npx skills add https://github.com/shushuzn/Rairos --skill geniml-shushuzn
Geniml turns genomic interval data (BED files) into usable machine-learning representations, so you can search, cluster, and model similarities among regions, labels, and single cells without manual feature engineering.
Use geniml to train Region2Vec embeddings by first tokenizing your BED files against a universe reference and then running Region2Vec training with the produced token folder.
High-intent search queries and answers about installing and using this skill.
To generate embeddings from BED files, you first tokenize your BED collections against a consensus peak universe, then train unsupervised Region2Vec models to produce vector representations for bulk region similarity and downstream ML tasks.
Yes, you can generate cell-level embeddings from scATAC-seq peak matrices using the scEmbed approach, which applies Region2Vec-style token training to support clustering, visualization, and cell annotation workflows.
A consensus peak universe is a standardized reference built from many experiments using CC, CCF, ML, or HMM methods. It is required to tokenize genomic interval data so that BED files map to a consistent vocabulary for embedding training.
You can perform metadata-aware similarity search by training BEDspace joint embeddings, which create a shared embedding space across region sets and metadata labels to enable region-to-label and label-to-region queries.
Yes, training Region2Vec requires a predefined universe reference. You must tokenize your BED files against this BED-derived universe to generate the token folder needed for the subsequent embedding training and inference flows.
Available utilities include caching, tokenization strategies, null-model randomization for statistical validation, embedding evaluation metrics, and neural search backends to index and query the trained genomic interval representations.
LeonChaoXBuild machine learning embeddings from genomic interval BED files.
hxk622Train genomic region embeddings and analyze single-cell ATAC-seq data with Python.
jimmc414Train Region2Vec embeddings on genomic interval data from BED files.
jackspaceGenerate unsupervised embeddings for genomic intervals from BED files.
