GenoME is a Mixture of Experts (MoE)-based generative model that integrates DNA sequence and cell-type-specific chromatin accessibility (ATAC-seq/DNase-seq) to predict a unified genomic profile across multiple scales and modalities. It enables individualized, multimodal prediction and perturbation of genomic profiles.
Paper: bioRxiv Preprint | Demo Data: Data link
- Multi-modal Prediction: Multimodal prediction of epigenomics, transcriptomics, and 3D chromatin architecture at base-pair to kilobase resolutions
- Cross-Cell Generalization: Cross-cell-type generalization to predict full regulatory landscapes for unseen or individualized cell types
- Perturbation Analysis: In silico perturbation analysis for simulating genetic and epigenetic perturbations and identifying functional regulatory connections
- The simplest way to set up the environment is using the provided
environment.ymlfile.git clone https://github.com/JWei2015/GenoME.git cd GenoME conda env create -f environment.yml conda activate py310 pip install -e .
- If the quick install does not work (e.g., due to dependency conflicts), follow these steps:
- Clone the repository
git clone https://github.com/JWei2015/GenoME.git cd GenoME - Create a fresh conda environment
Python>=3.10 is required; setuptools < 82.0.0 to avoid code conflicts.
conda create -n py310 python==3.10.20 setuptools==80.9.0 conda activate py310
- Install core dependencies
conda install mamba==2.5.0 mamba install pytorch==2.10.0 mamba install mamba-ssm==2.3.0 fsspec==2026.2.0 pip install cooltools mamba install cooler pip install pytorch-lightning==1.9.5 lightning-bolts==0.7.0 torch==2.10.0 mamba install kipoiseq mamba install pybigwig
- Input Formats:
- DNA sequence: FASTA format (hg38 reference genome)
- ATAC-seq/DNase-seq: BAM format (base-pair resolution)
- Training targets: BigWig files for RNA-seq, ChIP-seq; cooler format for Hi-C
- Data preprocessing： see Paper: BioRxiv Preprint