Skip to content
ModelVerifiedExternal
Download

Encode Chrombpnet DNASE ENCSR277KRY ENCSR167VCE

As part of the ENCODE 4 Project, we trained ChromBPNet models on 1,512 ENCODE DNAse-seq and ATAC-seq across 408 biosamples. Here, we provide all models for open-source use.

Sourcehuggingface.co/kundajelab/encode-chrombpnet-DNASE-ENCSR277KRY-ENCSR167VCE/tree/10be8c90c6fb3cca01dfdaebfeed0d3f804de99b
KindOtherVersionRevision 10be8c9Publisher@kundajelabCgrade

Model source

Kind
Other
File size
2.56–25.2 MB
Version
vRevision 10be8c9
Updated
Sep 22, 2026
Publisher
@kundajelab
Downloads
0
Source
Hugging Face

Source description

ENCODE ChromBPNet Atlas

As part of the ENCODE 4 Project, we trained ChromBPNet models on 1,512 ENCODE DNAse-seq and ATAC-seq across 408 biosamples. Here, we provide all models for open-source use.

For more information about the models, see:

ChromBPNet model: DNASE in adrenal gland (ENCSR277KRY)

  • Model: ChromBPNet
  • Assay: DNASE-seq
  • Experiment: ENCSR277KRY
  • Model annotation: ENCSR167VCE
  • Biosample: adrenal gland (Full name: Homo sapiens adrenal gland tissue male adult (37 years))
  • Cell slim(s): None
  • Organ slim(s): adrenal-gland,endocrine-gland
  • Developmental slim(s): mesoderm,ectoderm
  • System slim(s): endocrine-system
  • Assembly: hg38

Directory structure

  • fold_0: Model of 5-fold cross-validation: Fold 0
    • model.chrombpnet.fold_0.encid.h5: full chrombpnet model that combines both bias and corrected model in .h5 format
    • model.chrombpnet_nobias.fold_0.encid.h5: bias-corrected accessibility model in .h5 format (Use for all biological discovery)
    • model.bias_scaled.fold_0.encid.h5: bias model in .h5 format
    • model.chrombpnet.fold_0.encid.tar: full chrombpnet model that combines both bias and corrected model in SavedModel format. After being untarred, it results in a directory named "chrombpnet".
    • model.chrombpnet_nobias.fold_0.encid.tar: bias-corrected accessibility model in SavedModel format (Use for all biological discovery). After being untarred, it results in a directory named "chrombpnet_wo_bias".
    • model.bias_scaled.fold_0.encid.tar: bias model in SavedModel format. After being untarred, it results in a directory named "bias_model_scaled".
    • logs.models.fold_0.encid: folder containing log files for training models
  • fold_1: Model of 5-fold coss-validation: Fold 1
  • fold_2: Model of 5-fold cross-validation: Fold 2
  • fold_3: Model of 5-fold cross-validation: Fold 3
  • fold_4: Model of 5-fold cross-validation: Fold 4

Instructions

1. Pseudocode for loading models in .h5 format

(1) Use the code in python after appropriately defining model_in_h5_format and inputs.
(2) inputs is a one hot encoded sequence of shape (N,2114,4). Here N corresponds to the number of tested sequences, 2114 is the input sequence length and 4 corresponds to [A,C,G,T].

import tensorflow as tf
from tensorflow.keras.utils import get_custom_objects
from tensorflow.keras.models import load_model

custom_objects={"tf": tf}
get_custom_objects().update(custom_objects)

model=load_model(model_in_h5_format,compile=False)
outputs = model(inputs)

The list outputs consists of two elements. The first element has a shape of (N, 1000) and contains logit predictions for a 1000-base-pair output. The second element, with a shape of (N, 1), contains logcount predictions. To transform these predictions into per-base signals, follow the provided pseudo code lines below.

import numpy as np

def softmax(x, temp=1):
    norm_x = x - np.mean(x,axis=1, keepdims=True)
    return np.exp(temp*norm_x)/np.sum(np.exp(temp*norm_x), axis=1, keepdims=True)
    
predictions = softmax(outputs[0]) * (np.exp(outputs[1])-1)

2. Pseudocode for loading models in .tar format

(1) First untar the directory as follows tar -xvf model.tar.
(2) Use the code below in python after appropriately defining model_dir_untared and inputs.
(3) inputs is a one hot encoded sequence of shape (N,2114,4). Here N corresponds to the number of tested sequences, 2114 is the input sequence length and 4 corresponds to ACGT.

Reference: https://www.tensorflow.org/api_docs/python/tf/saved_model/load

import tensorflow as tf

model = tf.saved_model.load('model_dir_untared')
outputs = model.signatures['serving_default'](**{'sequence':inputs.astype('float32')})

The variable outputs represents a dictionary containing two key-value pairs. The first key is logits_profile_predictions, holding a value with a shape of (N, 1000). This value corresponds to logit predictions for a 1000-base-pair output. The second key, named `logcount_predictions``, is associated with a value of shape (N, 1), representing logcount predictions. To transform these predictions into per-base signals, utilize the provided pseudo code lines mentioned below.

import numpy as np
def softmax(x, temp=1):
    norm_x = x - np.mean(x,axis=1, keepdims=True)
    return np.exp(temp*norm_x)/np.sum(np.exp(temp*norm_x), axis=1, keepdims=True)
    
predictions = softmax(outputs["logits_profile_predictions"]) * (np.exp(outputs["logcount_predictions"])-1)

Docker image to load and use the models

Code for ChromBPNet

License & citation

External data users may freely download, analyze and publish results based on any ENCODE data without restrictions.

Released under the ENCODE data-use policy. Please cite the ENCODE Project Consortium and the model software: ChromBPNet (Pampari et al., bioRxiv 2024).

Sources

1 source
Hugging FacePrimaryVersion 10be8c90c6fb3cca01dfdaebfeed0d3f804de99b
huggingface.co/kundajelab/encode-chrombpnet-DNASE-ENCSR277KRY-ENCSR167VCE/tree/10be8c90c6fb3cca01dfdaebfeed0d3f804de99b

Verified Aug 6

Model artifacts

15 artifacts

fold_0/model.bias_scaled.fold_0.ENCSR277KRY.h5

h5 · 2.56 MB · SHA-256 1fc610698ead…0631 · Hugging Face

Download

fold_0/model.chrombpnet_nobias.fold_0.ENCSR277KRY.h5

h5 · 24.4 MB · SHA-256 c300c1c31349…f617 · Hugging Face

Download

fold_0/model.chrombpnet.fold_0.ENCSR277KRY.h5

h5 · 25.2 MB · SHA-256 79460c7f3f90…c286 · Hugging Face

Download

fold_1/model.bias_scaled.fold_1.ENCSR277KRY.h5

h5 · 2.56 MB · SHA-256 ca8ebbc649bb…c585 · Hugging Face

Download

fold_1/model.chrombpnet_nobias.fold_1.ENCSR277KRY.h5

h5 · 24.4 MB · SHA-256 13ef818c4ab8…e877 · Hugging Face

Download

fold_1/model.chrombpnet.fold_1.ENCSR277KRY.h5

h5 · 25.2 MB · SHA-256 34ed5d5ed19c…18dc · Hugging Face

Download

fold_2/model.bias_scaled.fold_2.ENCSR277KRY.h5

h5 · 2.56 MB · SHA-256 662e4a1dd755…f405 · Hugging Face

Download

fold_2/model.chrombpnet_nobias.fold_2.ENCSR277KRY.h5

h5 · 24.4 MB · SHA-256 602821099ad1…3da8 · Hugging Face

Download

fold_2/model.chrombpnet.fold_2.ENCSR277KRY.h5

h5 · 25.2 MB · SHA-256 de3b5a2c0445…40d4 · Hugging Face

Download

fold_3/model.bias_scaled.fold_3.ENCSR277KRY.h5

h5 · 2.56 MB · SHA-256 cd52611ca0e1…8661 · Hugging Face

Download

fold_3/model.chrombpnet_nobias.fold_3.ENCSR277KRY.h5

h5 · 24.4 MB · SHA-256 5bae8f64acd6…d558 · Hugging Face

Download

fold_3/model.chrombpnet.fold_3.ENCSR277KRY.h5

h5 · 25.2 MB · SHA-256 fcf11e4d417d…678a · Hugging Face

Download

fold_4/model.bias_scaled.fold_4.ENCSR277KRY.h5

h5 · 2.56 MB · SHA-256 1d4cfe60cc91…818d · Hugging Face

Download

fold_4/model.chrombpnet_nobias.fold_4.ENCSR277KRY.h5

h5 · 24.4 MB · SHA-256 57b13ee693ac…843d · Hugging Face

Download

fold_4/model.chrombpnet.fold_4.ENCSR277KRY.h5

h5 · 25.2 MB · SHA-256 e66dbd980e1f…6035 · Hugging Face

Download

Source excerpts

3 excerpts
ReadmeAug 6huggingface.co/kundajelab/encode-chrombpnet-DNASE-ENCSR277KRY-ENCSR167VCE

--- license: mit library_name: chrombpnet tags: - encode - chrombpnet - chromatin-accessibility - DNASE - adrenal - hg38 --- # ENCODE ChromBPNet Atlas As part of the ENCODE 4 Project, we trained ChromBPNet models on 1,512 ENCODE DNAse-seq and ATAC-seq across 408 biosamples. Here, we provide all models for open-source use. For more information about the models, see: - Main ENCODE 4 Paper - [A unified lexicon of predictive DNA sequence motifs from ENCODE transcription factor binding and chromatin accessibility assays](https://doi.org/10.5281/zenodo.17123347) (Yun, C. M. et al., Zenodo 2026) - [ChromBPNet: bias factorized, base-resolution deep learning models of chromatin accessibility reveal cis-regulatory sequence syntax, transcription factor footprints and regulatory variants](https://doi.org/10.1101/2024.12.25.630221) (Pampari, A. et al., bioRxiv 2024) ## ChromBPNet model: DNASE in adrenal gland (ENCSR277KRY) - Model: ChromBPNet - Assay: DNASE-seq - Experiment: [ENCSR277KRY](https://www.encodeproject.org/experiments/ENCSR277KRY/) - Model annotation: [ENCSR167VCE](https://www.encodeproject.org/annotations/ENCSR167VCE/) - Biosample: adrenal gland (Full name: Homo sapiens adrenal gland tissue male adult (37 years)) - Cell slim(s): None - Organ slim(s): adrenal-gland,endocrine-gland - Developmental slim(s): mesoderm,ectoderm - System slim(s): endocrine-system - Assembly: hg38 ## Directory structure - `fold_0`: Model of 5-fold cross-validation: Fold 0 - `model.chrombpnet.fold_0.encid.h5`: full chrombpnet model that combines both bias and corrected model in .h5 format - `model.chrombpnet_nobias.fold_0.encid.h5`: bias-corrected accessibility model in .h5 format (Use for all biological discovery) - `model.bias_scaled.fold_0.encid.h5`: bias model in .h5 format - `model.chrombpnet.fold_0.encid.tar`: full chrombpnet model that combines both bias and corrected model in SavedModel format. After being untarred, it results in a directory named "chrombpnet". - `model.chrombpnet_nobias.fold_0.encid.tar`: bias-corrected accessibility model in SavedModel format (Use for all biological discovery). After being untarred, it results in a directory named "chrombpnet_wo_bias". - `model.bias_scaled.fold_0.encid.tar`: bias model in SavedModel format. After being untarred, it results in a directory named "bias_model_scaled". - `logs.models.fold_0.encid`: folder cont...

ReadmeJul 11huggingface.co/kundajelab/encode-chrombpnet-DNASE-ENCSR277KRY-ENCSR167VCE

--- license: mit library_name: chrombpnet tags: - encode - chrombpnet - chromatin-accessibility - DNASE - adrenal - hg38 --- # ENCODE ChromBPNet Atlas As part of the ENCODE 4 Project, we trained ChromBPNet models on 1,512 ENCODE DNAse-seq and ATAC-seq across 408 biosamples. Here, we provide all models for open-source use. For more information about the models, see: - Main ENCODE 4 Paper - [A unified lexicon of predictive DNA sequence motifs from ENCODE transcription factor binding and chromatin accessibility assays](https://doi.org/10.5281/zenodo.17123347) (Deshpande et al., Zenodo 2025) - [ChromBPNet: bias factorized, base-resolution deep learning models of chromatin accessibility reveal cis-regulatory sequence syntax, transcription factor footprints and regulatory variants](https://doi.org/10.1101/2024.12.25.630221) (Pampari et al., bioRxiv 2024) ## ChromBPNet model: DNASE in adrenal gland (ENCSR277KRY) - Model: ChromBPNet - Assay: DNASE-seq - Experiment: [ENCSR277KRY](https://www.encodeproject.org/experiments/ENCSR277KRY/) - Model annotation: [ENCSR167VCE](https://www.encodeproject.org/annotations/ENCSR167VCE/) - Biosample: adrenal gland (Full name: Homo sapiens adrenal gland tissue male adult (37 years)) - Cell slim(s): None - Organ slim(s): adrenal-gland,endocrine-gland - Developmental slim(s): mesoderm,ectoderm - System slim(s): endocrine-system - Assembly: hg38 ## Directory structure - `fold_0`: Model of 5-fold cross-validation: Fold 0 - `model.chrombpnet.fold_0.encid.h5`: full chrombpnet model that combines both bias and corrected model in .h5 format - `model.chrombpnet_nobias.fold_0.encid.h5`: bias-corrected accessibility model in .h5 format (Use for all biological discovery) - `model.bias_scaled.fold_0.encid.h5`: bias model in .h5 format - `model.chrombpnet.fold_0.encid.tar`: full chrombpnet model that combines both bias and corrected model in SavedModel format. After being untarred, it results in a directory named "chrombpnet". - `model.chrombpnet_nobias.fold_0.encid.tar`: bias-corrected accessibility model in SavedModel format (Use for all biological discovery). After being untarred, it results in a directory named "chrombpnet_wo_bias". - `model.bias_scaled.fold_0.encid.tar`: bias model in SavedModel format. After being untarred, it results in a directory named "bias_model_scaled". - `logs.models.fold_0.encid`: folder containin...

NoteJul 11huggingface.co/kundajelab/encode-chrombpnet-DNASE-ENCSR277KRY-ENCSR167VCE

Source context: 0 downloads · 0 likes · Library chrombpnet · Repo kundajelab/encode-chrombpnet-DNASE-ENCSR277KRY-ENCSR167VCE