Upload folder using huggingface_hub

README.md

@@ -0,0 +1,120 @@
+---
+license: mit
+library_name: chrombpnet
+tags:
+  - encode
+  - chrombpnet
+  - chromatin-accessibility
+  - DNASE
+  - t-cell
+  - 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 stimulated activated CD8-positive, alpha-beta memory T cell (ENCSR749YCR)
+- Model: ChromBPNet
+- Assay: DNASE-seq
+- Experiment: [ENCSR749YCR](https://www.encodeproject.org/experiments/ENCSR749YCR/)
+- Model annotation: [ENCSR712YRU](https://www.encodeproject.org/annotations/ENCSR712YRU/)
+- Biosample: stimulated activated CD8-positive, alpha-beta memory T cell (Full name: Homo sapiens stimulated activated CD8-positive, alpha-beta memory T cell male adult (38 years) treated with anti-CD3 and anti-CD28 coated beads for 1 hour, 100 ng/mL Interleukin-7 for 1 hour)
+- Cell slim(s): T-cell,CD8+-T-cell,leukocyte,hematopoietic-cell
+- Organ slim(s): bodily-fluid,blood
+- Developmental slim(s): None
+- System slim(s): immune-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].
+
+```python
+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.
+
+```python
+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
+
+```python
+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.
+
+```python
+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
+- https://hub.docker.com/r/kundajelab/chrombpnet-atlas/ (tag:v1)
+
+## Code for ChromBPNet
+- https://github.com/kundajelab/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](https://www.encodeproject.org/about/data-use-policy/). Please cite the ENCODE Project Consortium and the model software: [ChromBPNet](https://github.com/kundajelab/chrombpnet) (Pampari et al., bioRxiv 2024).

fold_0/logs.models.fold_0.ENCSR749YCR/logfile.modelling.fold_0.ENCSR749YCR.args.json

@@ -0,0 +1,50 @@
+{
+    "cmd": "pipeline",
+    "genome": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_no_alt_analysis_set_GCA_000001405.15.fasta",
+    "chrom_sizes": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_EBV.chrom.sizes.tsv",
+    "input_bam_file": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR749YCR/preprocessing/bigWigs/ENCSR749YCR.bigWig",
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "output_dir": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_0",
+    "data_type": "DNASE",
+    "peaks": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_0/auxiliary/filtered.peaks.bed",
+    "nonpeaks": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_0/auxiliary/filtered.nonpeaks.bed",
+    "chr_fold_path": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_0.json",
+    "outlier_threshold": 0.9999,
+    "ATAC_ref_path": null,
+    "DNASE_ref_path": null,
+    "num_samples": 10000,
+    "inputlen": 2114,
+    "outputlen": 1000,
+    "seed": 1234,
+    "epochs": 50,
+    "early_stop": 5,
+    "learning_rate": 0.001,
+    "trackables": [
+        "logcount_predictions_loss",
+        "loss",
+        "logits_profile_predictions_loss",
+        "val_logcount_predictions_loss",
+        "val_loss",
+        "val_logits_profile_predictions_loss"
+    ],
+    "architecture_from_file": "/home/users/vhecht/chrombpnet/chrombpnet/chrombpnet/training/models/chrombpnet_with_bias_model.py",
+    "file_prefix": null,
+    "html_prefix": "./",
+    "bsort": false,
+    "tmpdir": null,
+    "no_st": false,
+    "bias_model_path": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/reference/ENCSR283TME_bias_0.9.h5",
+    "negative_sampling_ratio": 0.1,
+    "filters": 512,
+    "n_dilation_layers": 8,
+    "max_jitter": 500,
+    "batch_size": 64,
+    "output_prefix": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_0/models/chrombpnet",
+    "bigwig": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_0/auxiliary/data_unstranded.bw",
+    "plus_shift": null,
+    "minus_shift": null,
+    "chr": "chr8",
+    "pwm_width": 24,
+    "params": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_0/logs/chrombpnet_model_params.tsv"
+}

fold_0/logs.models.fold_0.ENCSR749YCR/logfile.modelling.fold_0.ENCSR749YCR.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_0/models/bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR749YCR/fold_0/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR749YCR/logfile.modelling.fold_0.ENCSR749YCR.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	1.0
+counts_sum_max_thresh	2413.66
+trainings_pts_post_thresh	165851

fold_0/logs.models.fold_0.ENCSR749YCR/logfile.modelling.fold_0.ENCSR749YCR.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_0/models/chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR749YCR/fold_0/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR749YCR/logfile.modelling.fold_0.ENCSR749YCR.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	1.9
+filters	512
+n_dil_layers	8
+bias_model_path	/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_0/models/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_0.json
+negative_sampling_ratio	0.1

fold_0/logs.models.fold_0.ENCSR749YCR/logfile.modelling.fold_0.ENCSR749YCR.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_0/models/chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR749YCR/fold_0/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR749YCR/logfile.modelling.fold_0.ENCSR749YCR.epoch_loss.csv

@@ -0,0 +1,14 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,1.2596428394317627,178.9869842529297,181.3800811767578,0.6270533800125122,162.31703186035156,163.5084991455078
+1,0.6140088438987732,170.61329650878906,171.7801055908203,0.4865908920764923,160.39830017089844,161.32275390625
+2,0.536813497543335,168.4860076904297,169.50628662109375,0.39730244874954224,158.4228973388672,159.17771911621094
+3,0.5011162757873535,167.0964813232422,168.04855346679688,0.3757857382297516,158.26522827148438,158.97914123535156
+4,0.4728296995162964,165.88589477539062,166.7838592529297,0.3772590160369873,157.4310302734375,158.1479034423828
+5,0.4606197476387024,164.86778259277344,165.7427520751953,0.37802818417549133,157.7228546142578,158.44100952148438
+6,0.44915279746055603,164.1803436279297,165.0337371826172,0.37192538380622864,156.92623901367188,157.63296508789062
+7,0.44020071625709534,163.4373321533203,164.2736053466797,0.40144917368888855,156.37588500976562,157.13861083984375
+8,0.42704299092292786,163.18695068359375,163.99815368652344,0.3928622603416443,158.05349731445312,158.7999267578125
+9,0.42312541604042053,162.20787048339844,163.01165771484375,0.39378565549850464,157.77264404296875,158.52088928222656
+10,0.4096429944038391,161.67288208007812,162.45126342773438,0.3581443727016449,157.1909942626953,157.87139892578125
+11,0.4113155007362366,161.57444763183594,162.3558807373047,0.3576602339744568,157.8583984375,158.53794860839844
+12,0.40316835045814514,161.09396362304688,161.8600311279297,0.3783148527145386,159.36582946777344,160.08453369140625

fold_0/model.bias_scaled.fold_0.ENCSR749YCR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc4d34ff5cdee3613f677035e6f83de2891a3c969ceb93676fffa2ae2c31a8dc
+size 2691928

fold_0/model.bias_scaled.fold_0.ENCSR749YCR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fd0374884cbc69d067a1a3869b58a4306511225e26db9f648178bca9e9436c4
+size 1198080

fold_0/model.chrombpnet.fold_0.ENCSR749YCR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fe783e02853b9dd8be00bb37b715c010b02f69e4bcc24766901d768446109c8
+size 77538896

fold_0/model.chrombpnet.fold_0.ENCSR749YCR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:901675625f3baa026e1e429b3356ff337e3caf9ad1c80237a54bcf2a20a3b7ce
+size 27535360

fold_0/model.chrombpnet_nobias.fold_0.ENCSR749YCR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c94bfe71bafd6ba5d2aadc8cb93c517c638bc77ddf5f843d2f86024f6a647397
+size 25582648

fold_0/model.chrombpnet_nobias.fold_0.ENCSR749YCR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2e655b00a711d3891c735a60f3b880283c4b2d45b53a5d68bda7cfe1f5697a4
+size 26060800

fold_1/logs.models.fold_1.ENCSR749YCR/logfile.modelling.fold_1.ENCSR749YCR.args.json

@@ -0,0 +1,50 @@
+{
+    "cmd": "pipeline",
+    "genome": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_no_alt_analysis_set_GCA_000001405.15.fasta",
+    "chrom_sizes": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_EBV.chrom.sizes.tsv",
+    "input_bam_file": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR749YCR/preprocessing/bigWigs/ENCSR749YCR.bigWig",
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "output_dir": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_1",
+    "data_type": "DNASE",
+    "peaks": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_1/auxiliary/filtered.peaks.bed",
+    "nonpeaks": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_1/auxiliary/filtered.nonpeaks.bed",
+    "chr_fold_path": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_1.json",
+    "outlier_threshold": 0.9999,
+    "ATAC_ref_path": null,
+    "DNASE_ref_path": null,
+    "num_samples": 10000,
+    "inputlen": 2114,
+    "outputlen": 1000,
+    "seed": 1234,
+    "epochs": 50,
+    "early_stop": 5,
+    "learning_rate": 0.001,
+    "trackables": [
+        "logcount_predictions_loss",
+        "loss",
+        "logits_profile_predictions_loss",
+        "val_logcount_predictions_loss",
+        "val_loss",
+        "val_logits_profile_predictions_loss"
+    ],
+    "architecture_from_file": "/home/users/vhecht/chrombpnet/chrombpnet/chrombpnet/training/models/chrombpnet_with_bias_model.py",
+    "file_prefix": null,
+    "html_prefix": "./",
+    "bsort": false,
+    "tmpdir": null,
+    "no_st": false,
+    "bias_model_path": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/reference/ENCSR283TME/fold_1/bias.h5",
+    "negative_sampling_ratio": 0.1,
+    "filters": 512,
+    "n_dilation_layers": 8,
+    "max_jitter": 500,
+    "batch_size": 64,
+    "output_prefix": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_1/models/chrombpnet",
+    "bigwig": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_1/auxiliary/data_unstranded.bw",
+    "plus_shift": null,
+    "minus_shift": null,
+    "chr": "chr12",
+    "pwm_width": 24,
+    "params": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_1/logs/chrombpnet_model_params.tsv"
+}

fold_1/logs.models.fold_1.ENCSR749YCR/logfile.modelling.fold_1.ENCSR749YCR.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_1/models/bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR749YCR/fold_1/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR749YCR/logfile.modelling.fold_1.ENCSR749YCR.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	1.0
+counts_sum_max_thresh	2419.2
+trainings_pts_post_thresh	167926

fold_1/logs.models.fold_1.ENCSR749YCR/logfile.modelling.fold_1.ENCSR749YCR.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_1/models/chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR749YCR/fold_1/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR749YCR/logfile.modelling.fold_1.ENCSR749YCR.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	1.9
+filters	512
+n_dil_layers	8
+bias_model_path	/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_1/models/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_1.json
+negative_sampling_ratio	0.1

fold_1/logs.models.fold_1.ENCSR749YCR/logfile.modelling.fold_1.ENCSR749YCR.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_1/models/chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR749YCR/fold_1/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR749YCR/logfile.modelling.fold_1.ENCSR749YCR.epoch_loss.csv

@@ -0,0 +1,15 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,1.6492770910263062,175.4113006591797,178.5449676513672,0.8591715693473816,213.1370086669922,214.76947021484375
+1,0.6107573509216309,166.84445190429688,168.00491333007812,0.5222768187522888,207.95567321777344,208.94796752929688
+2,0.548599362373352,165.0384979248047,166.08038330078125,0.4742248058319092,207.0544891357422,207.9555206298828
+3,0.50742107629776,163.4241180419922,164.38787841796875,0.5012868642807007,206.28515625,207.23773193359375
+4,0.4781603217124939,162.19041442871094,163.0990753173828,0.451394647359848,206.48912048339844,207.34666442871094
+5,0.464964359998703,161.6877899169922,162.57098388671875,0.44769883155822754,205.822998046875,206.6736602783203
+6,0.4456605613231659,160.8026885986328,161.649658203125,0.4484063386917114,204.92747497558594,205.77931213378906
+7,0.44236621260643005,160.19822692871094,161.0386199951172,0.4349081814289093,206.4998321533203,207.3262481689453
+8,0.42875242233276367,159.37551879882812,160.19017028808594,0.42513492703437805,204.6658172607422,205.47366333007812
+9,0.4172358512878418,158.5816192626953,159.3742218017578,0.4075920879840851,205.25791931152344,206.0324249267578
+10,0.4093433916568756,158.3081817626953,159.08628845214844,0.4185829162597656,205.89918518066406,206.6947021484375
+11,0.40561577677726746,157.8881072998047,158.65899658203125,0.3991405963897705,204.86056518554688,205.61898803710938
+12,0.3988059461116791,157.2968292236328,158.0546875,0.42720887064933777,205.07009887695312,205.88169860839844
+13,0.3927890360355377,156.83676147460938,157.5836639404297,0.4272770285606384,206.5803680419922,207.39219665527344

fold_1/model.bias_scaled.fold_1.ENCSR749YCR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5c7f1a22238a5c270e092a8ea88debd7da8d96695899785f58090de25e53831
+size 2691928

fold_1/model.bias_scaled.fold_1.ENCSR749YCR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82c082070c53df850013878946eb346848f03b4793167a672a89e8914dbdda86
+size 1198080

fold_1/model.chrombpnet.fold_1.ENCSR749YCR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3bef42b194e72e85a93ba955ca018abfc87ff4f2a7088b8c1aedbd7cc32b86c
+size 77538840

fold_1/model.chrombpnet.fold_1.ENCSR749YCR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f01bea99497e9e08d75d5d1e52f9a5a43ea51ff157c40b50b6fc338fc816e2f
+size 27525120

fold_1/model.chrombpnet_nobias.fold_1.ENCSR749YCR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:700b445b5843ac261cb89f11e7051723d7edaebce67e03a8d71db58a6e7c8e00
+size 25582648

fold_1/model.chrombpnet_nobias.fold_1.ENCSR749YCR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9609e2a7516a40f4c55549c5e95b9b95bbebc6dee0dcad1a220340f98e2c697
+size 26060800

fold_2/logs.models.fold_2.ENCSR749YCR/logfile.modelling.fold_2.ENCSR749YCR.args.json

@@ -0,0 +1,50 @@
+{
+    "cmd": "pipeline",
+    "genome": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_no_alt_analysis_set_GCA_000001405.15.fasta",
+    "chrom_sizes": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_EBV.chrom.sizes.tsv",
+    "input_bam_file": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR749YCR/preprocessing/bigWigs/ENCSR749YCR.bigWig",
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "output_dir": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_2",
+    "data_type": "DNASE",
+    "peaks": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_2/auxiliary/filtered.peaks.bed",
+    "nonpeaks": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_2/auxiliary/filtered.nonpeaks.bed",
+    "chr_fold_path": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_2.json",
+    "outlier_threshold": 0.9999,
+    "ATAC_ref_path": null,
+    "DNASE_ref_path": null,
+    "num_samples": 10000,
+    "inputlen": 2114,
+    "outputlen": 1000,
+    "seed": 1234,
+    "epochs": 50,
+    "early_stop": 5,
+    "learning_rate": 0.001,
+    "trackables": [
+        "logcount_predictions_loss",
+        "loss",
+        "logits_profile_predictions_loss",
+        "val_logcount_predictions_loss",
+        "val_loss",
+        "val_logits_profile_predictions_loss"
+    ],
+    "architecture_from_file": "/home/users/vhecht/chrombpnet/chrombpnet/chrombpnet/training/models/chrombpnet_with_bias_model.py",
+    "file_prefix": null,
+    "html_prefix": "./",
+    "bsort": false,
+    "tmpdir": null,
+    "no_st": false,
+    "bias_model_path": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/reference/ENCSR283TME/fold_2/bias.h5",
+    "negative_sampling_ratio": 0.1,
+    "filters": 512,
+    "n_dilation_layers": 8,
+    "max_jitter": 500,
+    "batch_size": 64,
+    "output_prefix": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_2/models/chrombpnet",
+    "bigwig": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_2/auxiliary/data_unstranded.bw",
+    "plus_shift": null,
+    "minus_shift": null,
+    "chr": "chr22",
+    "pwm_width": 24,
+    "params": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_2/logs/chrombpnet_model_params.tsv"
+}

fold_2/logs.models.fold_2.ENCSR749YCR/logfile.modelling.fold_2.ENCSR749YCR.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_2/models/bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR749YCR/fold_2/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR749YCR/logfile.modelling.fold_2.ENCSR749YCR.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	1.0
+counts_sum_max_thresh	2445.3
+trainings_pts_post_thresh	173545

fold_2/logs.models.fold_2.ENCSR749YCR/logfile.modelling.fold_2.ENCSR749YCR.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_2/models/chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR749YCR/fold_2/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR749YCR/logfile.modelling.fold_2.ENCSR749YCR.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	1.9
+filters	512
+n_dil_layers	8
+bias_model_path	/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_2/models/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_2.json
+negative_sampling_ratio	0.1

fold_2/logs.models.fold_2.ENCSR749YCR/logfile.modelling.fold_2.ENCSR749YCR.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_2/models/chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR749YCR/fold_2/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR749YCR/logfile.modelling.fold_2.ENCSR749YCR.epoch_loss.csv

@@ -0,0 +1,12 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,1.6126301288604736,177.4844512939453,180.54849243164062,0.6261447668075562,178.41616821289062,179.6057891845703
+1,0.597298264503479,168.70620727539062,169.84097290039062,0.47797319293022156,176.50830078125,177.41641235351562
+2,0.5378792881965637,167.03445434570312,168.05642700195312,0.4528815746307373,174.23675537109375,175.09725952148438
+3,0.500616192817688,165.70794677734375,166.6591033935547,0.4252174198627472,174.7831573486328,175.59115600585938
+4,0.4843604266643524,164.58619689941406,165.5065460205078,0.45756420493125916,174.1162872314453,174.98568725585938
+5,0.45230287313461304,163.87448120117188,164.7337188720703,0.41353198885917664,173.28675842285156,174.07247924804688
+6,0.44713664054870605,163.22012329101562,164.0697479248047,0.4163692891597748,173.6237335205078,174.41481018066406
+7,0.4366699457168579,162.54989624023438,163.37989807128906,0.3991748094558716,174.15988159179688,174.91818237304688
+8,0.424078106880188,161.6614990234375,162.46737670898438,0.39096349477767944,174.19973754882812,174.94248962402344
+9,0.4176090955734253,161.17007446289062,161.9638671875,0.4565593898296356,174.0085906982422,174.87600708007812
+10,0.40975940227508545,160.6403045654297,161.41883850097656,0.39465779066085815,174.69790649414062,175.44757080078125

fold_2/model.bias_scaled.fold_2.ENCSR749YCR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95ce57e21e8548db643a0c5a39800b789ddb76757e7beb38b40917eb6248b578
+size 2691928

fold_2/model.bias_scaled.fold_2.ENCSR749YCR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36c3872df7bfe68f45d33980ac39fc088057913f877a71829de1fd5c5ff451ff
+size 1198080

fold_2/model.chrombpnet.fold_2.ENCSR749YCR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64019731b8b551d3fbd9757d29f89687cd3458353576770b40f4c3eda2a64783
+size 77538840

fold_2/model.chrombpnet.fold_2.ENCSR749YCR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75410933d88612350a3cbebeed75c6365058ae641bd80b6eaa0ed93406e171c8
+size 27525120

fold_2/model.chrombpnet_nobias.fold_2.ENCSR749YCR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1bd6db23f0ee959347dc6170c9c85381097b7a8c3becdb47d9bbdd8296aa225c
+size 25582648

fold_2/model.chrombpnet_nobias.fold_2.ENCSR749YCR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ac24cedefab646969ebc73d3a2f1ac7fa0b2795b296c89a6681dd45cba3752d
+size 26060800

fold_3/logs.models.fold_3.ENCSR749YCR/logfile.modelling.fold_3.ENCSR749YCR.args.json

@@ -0,0 +1,50 @@
+{
+    "cmd": "pipeline",
+    "genome": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_no_alt_analysis_set_GCA_000001405.15.fasta",
+    "chrom_sizes": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_EBV.chrom.sizes.tsv",
+    "input_bam_file": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR749YCR/preprocessing/bigWigs/ENCSR749YCR.bigWig",
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "output_dir": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_3",
+    "data_type": "DNASE",
+    "peaks": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_3/auxiliary/filtered.peaks.bed",
+    "nonpeaks": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_3/auxiliary/filtered.nonpeaks.bed",
+    "chr_fold_path": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_3.json",
+    "outlier_threshold": 0.9999,
+    "ATAC_ref_path": null,
+    "DNASE_ref_path": null,
+    "num_samples": 10000,
+    "inputlen": 2114,
+    "outputlen": 1000,
+    "seed": 1234,
+    "epochs": 50,
+    "early_stop": 5,
+    "learning_rate": 0.001,
+    "trackables": [
+        "logcount_predictions_loss",
+        "loss",
+        "logits_profile_predictions_loss",
+        "val_logcount_predictions_loss",
+        "val_loss",
+        "val_logits_profile_predictions_loss"
+    ],
+    "architecture_from_file": "/home/users/vhecht/chrombpnet/chrombpnet/chrombpnet/training/models/chrombpnet_with_bias_model.py",
+    "file_prefix": null,
+    "html_prefix": "./",
+    "bsort": false,
+    "tmpdir": null,
+    "no_st": false,
+    "bias_model_path": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/reference/ENCSR283TME/fold_3/bias.h5",
+    "negative_sampling_ratio": 0.1,
+    "filters": 512,
+    "n_dilation_layers": 8,
+    "max_jitter": 500,
+    "batch_size": 64,
+    "output_prefix": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_3/models/chrombpnet",
+    "bigwig": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_3/auxiliary/data_unstranded.bw",
+    "plus_shift": null,
+    "minus_shift": null,
+    "chr": "chr6",
+    "pwm_width": 24,
+    "params": "/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_3/logs/chrombpnet_model_params.tsv"
+}

fold_3/logs.models.fold_3.ENCSR749YCR/logfile.modelling.fold_3.ENCSR749YCR.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_3/models/bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR749YCR/fold_3/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_3/logs.models.fold_3.ENCSR749YCR/logfile.modelling.fold_3.ENCSR749YCR.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	1.0
+counts_sum_max_thresh	2442.21
+trainings_pts_post_thresh	166104

fold_3/logs.models.fold_3.ENCSR749YCR/logfile.modelling.fold_3.ENCSR749YCR.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_3/models/chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR749YCR/fold_3/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_3/logs.models.fold_3.ENCSR749YCR/logfile.modelling.fold_3.ENCSR749YCR.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	1.9
+filters	512
+n_dil_layers	8
+bias_model_path	/oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_3/models/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_3.json
+negative_sampling_ratio	0.1

fold_3/logs.models.fold_3.ENCSR749YCR/logfile.modelling.fold_3.ENCSR749YCR.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin_atlas_bias_corrected/DNASE_model/ENCSR749YCR/fold_3/models/chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR749YCR/fold_3/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py