SkillAgentSearch skills...

geniml

Python library for genomic interval ML. Train/apply region2vec embeddings turning BED regions into vectors, index interval datasets for ML, search embedding space with BEDSpace, and evaluate embedding quality.

Install / Use

npx skills add jaechang-hits/SciAgent-Skills --skill geniml

Installs into whichever agent you are using.

About this skill
📄

SKILL.md

Installable skill definition

Quality Score

91/100

Supported Platforms

Universal

Our assessment of geniml

geniml scores 91/100 on our quality scale, 347th of 944 AI & Machine Learning skills we index (top 37%).

Its SKILL.md is 17 KB long, well organised into 65 sections with 18 code examples: a thorough specification that gives an agent plenty to work with.

It has 367 GitHub stars, a meaningful sign that others use it.

Substance
30/30
Structure
20/20
Description
15/15
Adoption
11/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated about 3 months ago, so geniml is actively maintained.
  • Our last check on 2026-11-20 found the source still online.
  • No license is declared. By default that means all rights are reserved: you can read it, but reusing or redistributing it is not clearly permitted. Ask the author before building on it commercially.
  • Its trust signals score 88/100, with 1 caution from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.

AI review by kimi-k2.7-code on 2026-10-05. Automated pattern scan on 2026-10-05. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

geniml compared with similar skills

All 4 of these similar skills score higher than geniml; compare them before choosing.

SkillScoreStarsUpdatedFormat
geniml (this skill)by jaechang-hits913673mo agoSKILL.md
claude-memby thedotmack100118.5ktodayCLAUDE.md
Agent-Reachby Panniantong100113.3k12d agoCLAUDE.md
Understand-Anythingby Egonex-AI10093.3k1d agoCLAUDE.md
headroomby headroomlabs-ai10080.2ktodayCLAUDE.md

Frequently asked questions

How do I install geniml?
Run npx skills add jaechang-hits/SciAgent-Skills --skill geniml. The install tabs above show the steps for each supported agent.
Which AI agents does geniml work with?
It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
Is geniml safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It declares no license and scores 88/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is geniml still maintained?
The repository was last updated about 3 months ago, so geniml is actively maintained.

name: "geniml" description: "Python library for genomic interval ML. Train/apply region2vec embeddings turning BED regions into vectors, index interval datasets for ML, search embedding space with BEDSpace, and evaluate embedding quality. Use for chromatin accessibility clustering, regulatory element classification, and cross-sample region comparison." license: "BSD-2-Clause"

Geniml: Genomic Interval Machine Learning

Overview

Geniml is a Python library that bridges genomic interval biology and machine learning. It provides region2vec for learning dense vector representations of genomic regions from BED files, BEDSpace for nearest-neighbor search in embedding space, dataset classes for ML-ready genomic interval loading, and evaluation utilities for embedding quality. Geniml is designed for researchers who want to apply modern ML techniques to chromatin accessibility, histone modification, or other region-based genomic data.

When to Use

  • Learn dense embeddings of genomic regions from a collection of BED files to enable ML-based analysis (region2vec)
  • Cluster chromatin accessibility peaks or histone modification sites by embedding similarity
  • Search for genomic regions similar to a query region using approximate nearest-neighbor search (BEDSpace)
  • Build training datasets for ML models from BED-format genomic intervals with a PyTorch-compatible interface
  • Compare embedding quality across training runs or datasets using quantitative metrics
  • Integrate genomic region representations into custom neural network architectures
  • For basic BED file parsing and set operations without ML, use gtars or pysam-genomic-files instead

Prerequisites

  • Python packages: geniml, torch, numpy, pandas, anndata
  • Data requirements: BED files (minimum 3 columns: chr, start, end); optionally a pre-built universe file
  • Environment: Python 3.8+; GPU optional but recommended for region2vec training on large datasets
pip install geniml torch numpy pandas anndata

Quick Start

from geniml.region2vec import Region2VecExModel
from geniml.io import RegionSet

# Load a collection of BED files and train region2vec embeddings
region_sets = [RegionSet("sample1.bed"), RegionSet("sample2.bed"), RegionSet("sample3.bed")]
model = Region2VecExModel("path/to/universe.bed")
model.train(region_sets, epochs=10, batch_size=32)

# Get embedding for a specific region
embedding = model.encode("chr1", 1000000, 1500000)
print(f"Embedding shape: {embedding.shape}")
# Embedding shape: (100,)

Core API

Module 1: RegionSet — Genomic Interval I/O

Load BED files into geniml's primary data structure for downstream operations.

from geniml.io import RegionSet

# Load a BED file
rs = RegionSet("peaks.bed")
print(f"Loaded {len(rs)} regions")
print(f"First region: {rs[0]}")          # Region object: chr, start, end
print(f"Chromosomes: {set(r.chr for r in rs)}")

# Convert to list of Region objects
regions = list(rs)
for r in regions[:3]:
    print(f"  {r.chr}:{r.start}-{r.end}")
from geniml.io import RegionSet

# Create RegionSet from a list of (chr, start, end) tuples
regions_data = [
    ("chr1", 100000, 101000),
    ("chr1", 200000, 201500),
    ("chr2",  50000,  51200),
]
rs = RegionSet(regions_data)
print(f"RegionSet with {len(rs)} regions from list")

# Access by index
r = rs[0]
print(f"chr={r.chr}, start={r.start}, end={r.end}, width={r.end - r.start}")

Module 2: Universe Building

A universe defines the set of consensus regions used as the vocabulary for region2vec. Build it from a collection of BED files.

from geniml.universe import UniverseBuilder
from geniml.io import RegionSet

# Collect BED files representing diverse samples
bed_files = ["sample1.bed", "sample2.bed", "sample3.bed", "sample4.bed"]
region_sets = [RegionSet(f) for f in bed_files]

# Build universe (consensus non-overlapping regions)
builder = UniverseBuilder()
universe = builder.build(region_sets)

# Save universe to BED file
universe.to_bed("universe.bed")
print(f"Universe size: {len(universe)} consensus regions")
from geniml.universe import UniverseBuilder
from geniml.io import RegionSet

# Build universe with custom parameters
builder = UniverseBuilder(
    fraction=0.5,    # Region must appear in >= 50% of samples to be included
    merge_dist=0,    # Merge adjacent regions within this distance (bp)
)
bed_files = [f"sample_{i}.bed" for i in range(1, 11)]
region_sets = [RegionSet(f) for f in bed_files]
universe = builder.build(region_sets)
print(f"Filtered universe: {len(universe)} regions (fraction >= 0.5)")

Module 3: Region2Vec — Training Embeddings

Train word2vec-style embeddings on genomic regions, treating each BED file as a "document" and each region as a "word."

from geniml.region2vec import Region2VecExModel
from geniml.io import RegionSet

# Initialize model with a pre-built universe
model = Region2VecExModel(universe="universe.bed", embedding_dim=100)

# Load training data (collection of BED files = corpus)
bed_files = [f"atac_{i}.bed" for i in range(1, 51)]
region_sets = [RegionSet(f) for f in bed_files]

# Train
model.train(
    region_sets,
    epochs=20,
    batch_size=64,
    window_size=5,
    min_count=1,
)
print("Training complete")

# Save trained model
model.save("region2vec_model/")
print("Model saved to region2vec_model/")
from geniml.region2vec import Region2VecExModel

# Load a pre-trained model
model = Region2VecExModel.load("region2vec_model/")

# Encode a single genomic region
embedding = model.encode("chr1", 1_000_000, 1_500_000)
print(f"Single region embedding shape: {embedding.shape}")
# Single region embedding shape: (100,)

# Encode an entire BED file → matrix of region embeddings
from geniml.io import RegionSet
rs = RegionSet("query_peaks.bed")
embeddings = model.encode_region_set(rs)
print(f"BED file embeddings shape: {embeddings.shape}")
# BED file embeddings shape: (N_regions, 100)

Module 4: BEDSpace — Embedding Nearest-Neighbor Search

Index a corpus of BED file embeddings for fast similarity search.

from geniml.bedspace import BEDSpace
from geniml.region2vec import Region2VecExModel
from geniml.io import RegionSet

# Build a BEDSpace index from a set of BED files
model = Region2VecExModel.load("region2vec_model/")
bed_files = [f"dataset_{i}.bed" for i in range(1, 101)]
region_sets = [RegionSet(f) for f in bed_files]

bedspace = BEDSpace(model)
bedspace.fit(region_sets, labels=[f"dataset_{i}" for i in range(1, 101)])
bedspace.save("bedspace_index/")
print(f"BEDSpace index built with {len(region_sets)} datasets")
from geniml.bedspace import BEDSpace
from geniml.io import RegionSet

# Load index and query
bedspace = BEDSpace.load("bedspace_index/")
query = RegionSet("query_sample.bed")

# Find top-k most similar datasets
results = bedspace.query(query, k=5)
for rank, (label, score) in enumerate(results, 1):
    print(f"  Rank {rank}: {label}  similarity={score:.4f}")

Module 5: Genomic Interval Datasets for ML

PyTorch-compatible Dataset classes for training ML models on genomic intervals.

from geniml.datasets import TokenizedBEDDataset
from torch.utils.data import DataLoader

# Create a tokenized dataset from multiple BED files + labels
bed_files = ["condition_A_rep1.bed", "condition_A_rep2.bed",
             "condition_B_rep1.bed", "condition_B_rep2.bed"]
labels   = [0, 0, 1, 1]

dataset = TokenizedBEDDataset(
    bed_files=bed_files,
    universe="universe.bed",
    labels=labels,
)
print(f"Dataset size: {len(dataset)} samples")

# Use with PyTorch DataLoader
loader = DataLoader(dataset, batch_size=8, shuffle=True)
for batch_tokens, batch_labels in loader:
    print(f"Batch tokens shape: {batch_tokens.shape}")
    print(f"Batch labels: {batch_labels}")
    break
from geniml.datasets import RegionEmbeddingDataset
from geniml.region2vec import Region2VecExModel
import torch

# Pre-embed BED files and create a dataset for a classifier
model = Region2VecExModel.load("region2vec_model/")
bed_files = ["pos_1.bed", "pos_2.bed", "neg_1.bed", "neg_2.bed"]
labels    = [1, 1, 0, 0]

emb_dataset = RegionEmbeddingDataset(
    bed_files=bed_files,
    model=model,
    labels=labels,
    aggregation="mean",    # aggregate region embeddings per sample
)
X, y = emb_dataset[0]
print(f"Sample embedding shape: {X.shape}, label: {y}")
# Sample embedding shape: (100,), label: 1

Module 6: Embedding Evaluation

Assess embedding quality using neighborhood overlap and intrinsic metrics.

from geniml.eval import EmbeddingEvaluator
from geniml.region2vec import Region2VecExModel
from geniml.io import RegionSet

model = Region2VecExModel.load("region2vec_model/")
bed_files = [f"sample_{i}.bed" for i in range(1, 21)]
region_sets = [RegionSet(f) for f in bed_files]

evaluator = EmbeddingEvaluator(model)
metrics = evaluator.evaluate(region_sets)

print(f"Neighborhood overlap score: {metrics['neighborhood_overlap']:.4f}")
print(f"Silhouette score:           {metrics.get('silhouette', 'N/A')}")
# Higher neighborhood overlap → more biologically coherent embeddings

Key Concepts

region2vec Training Analogy

Region2vec treats each BED file as a "sentence" (a document of co-occurring genomic regions) and each genomic region token (from the universe) as a "word." Word2vec's skip-gram objective is applied: regions that frequently co-occur in BED files are learned to have similar embeddings. This means two regions that tend to be open/active in the same set of samples will have nearby embeddings, enabling meaningful similarity search without explicit labels.

# Conceptual analogy: universe regions = vocabulary tokens
# BED file = sentence = [region_token_1, region_token_2, ...]
# Training: predict co-occurring regions within a sliding window
from geniml.io import RegionSet
rs = RegionSet("sample.bed")
print(f"This BED file contains {len(rs)} 'words' (tokens) from the universe vocabulary")

Universe as Vocabulary

The universe is a non-overlapping, genome-wide set of consensus regions that serves as the token vocabulary for region2vec. A well-chosen universe should cover the genomic regions present in your dataset while being compact enough for efficient training. Regions in a BED file that do not overlap the universe are ignored during training.

Common Workflows

Workflow 1: Train and Search Embeddings from ATAC-seq Peaks

Goal: Embed ATAC-seq peak sets, then find samples most similar to a query.

from geniml.universe import UniverseBuilder
from geniml.region2vec import Region2VecExModel
from geniml.bedspace import BEDSpace
from geniml.io import RegionSet
from pathlib import Path

# Step 1: Collect BED files
bed_dir = Path("atac_peaks/")
bed_files = sorted(bed_dir.glob("*.bed"))
region_sets = [RegionSet(str(f)) for f in bed_files]
labels = [f.stem for f in bed_files]
print(f"Loaded {len(region_sets)} ATAC-seq peak sets")

# Step 2: Build universe
builder = UniverseBuilder(fraction=0.3)
universe = builder.build(region_sets)
universe.to_bed("atac_universe.bed")
print(f"Universe: {len(universe)} regions")

# Step 3: Train region2vec
model = Region2VecExModel(universe="atac_universe.bed", embedding_dim=100)
model.train(region_sets, epochs=15, batch_size=64, window_size=5)
model.save("atac_region2vec/")

# Step 4: Build BEDSpace index
bedspace = BEDSpace(model)
bedspace.fit(region_sets, labels=labels)
bedspace.save("atac_bedspace/")

# Step 5: Query with a new sample
query = RegionSet("new_sample.bed")
results = bedspace.query(query, k=5)
print("Top 5 most similar samples:")
for rank, (label, score) in enumerate(results, 1):
    print(f"  {rank}. {label}  (similarity={score:.4f})")

Workflow 2: ML Classification on Genomic Intervals

Goal: Train a logistic regression classifier on region2vec embeddings to classify cell typ

Truncated for display — read the full file on GitHub.

Related Skills

View on GitHub
GitHub Stars367
CategoryAI
Updated2mo ago
Forks36

Languages

Python

Trust signals

88/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

1 medium
geniml — Universal Skill: Install & Safety Check | SkillAgent