ChimeraLM

A Genomic Language Model for Detecting WGA Chimeric Artifacts

Installation • Quick Start • Documentation • Citation

A deep learning-powered tool to identify chimeric artifacts introduced by whole genome amplification (WGA).

pip install chimeralm

Requirements: Python 3.10, 3.11 and 3.12

For GPU support, installation instructions, and troubleshooting, see the Installation Guide.

# Predict chimeric reads (CPU)
chimeralm predict your_data.bam

# Predict with GPU acceleration
chimeralm predict your_data.bam --gpus 1 --batch-size 24

# Filter BAM to remove chimeric reads
chimeralm filter your_data.bam your_data.predictions

Output:

• Predictions: Tab-separated file with read names and labels (0=biological, 1=chimeric)
• Filtered BAM: {input}.filtered.sorted.bam with chimeric reads removed

Need more help? See the Quick Start Tutorial for a complete walkthrough.

Full documentation is available at ylab-hi.github.io/ChimeraLM

Key Resources:

• Installation Guide - Setup with pip, conda, uv, or from source
• Quick Start Tutorial - Your first prediction in 15 minutes
• CLI Reference - Complete command documentation
• BAM Filtering Tutorial - Comprehensive filtering guide
• Performance Optimization - Speed up your analysis
• Troubleshooting - Common issues and solutions

• High Accuracy: Deep learning model trained on real WGA data
• GPU Accelerated: Optimized for CUDA, MPS (Apple Silicon), and CPU
• Easy to Use: Simple CLI with sensible defaults
• Fast Processing: Batch inference with configurable parallelism
• Web Interface: Interactive web UI for visualization and analysis
• Production Ready: Includes filtering, sorting, and indexing of BAM files

Contributions are welcome! See our Contributing Guide for development setup and guidelines.

If you use ChimeraLM in your research, please cite:

@software{chimeralm2025,
  title={ChimeraLM: A genomic language model to identify chimera artifacts},
  author={Li, Yangyang and Guo, Qingxiang and Yang, Rendong},
  year={2025},
  url={https://github.com/ylab-hi/ChimeraLM}
}

Apache License 2.0 - see LICENSE for details.