Upload folder using huggingface_hub

README.md

@@ -0,0 +1,120 @@
+---
+license: mit
+library_name: chrombpnet
+tags:
+  - encode
+  - chrombpnet
+  - chromatin-accessibility
+  - ATAC
+  - NCI-H929
+  - 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: ATAC in NCI-H929 (ENCSR382LBS)
+- Model: ChromBPNet
+- Assay: ATAC-seq
+- Experiment: [ENCSR382LBS](https://www.encodeproject.org/experiments/ENCSR382LBS/)
+- Model annotation: [ENCSR546DOB](https://www.encodeproject.org/annotations/ENCSR546DOB/)
+- Biosample: NCI-H929 (Full name: Homo sapiens NCI-H929)
+- Cell slim(s): cancer-cell,leukocyte,hematopoietic-cell,B-cell
+- Organ slim(s): bone-marrow,bone-element
+- Developmental slim(s): mesoderm
+- System slim(s): skeletal-system,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.ENCSR382LBS/logfile.modelling.fold_0.ENCSR382LBS.args.json

@@ -0,0 +1,50 @@
+{
+    "cmd": "pipeline",
+    "genome": "/oak/stanford/groups/akundaje/cmyun/chromatlas/reference/hg38.fa",
+    "chrom_sizes": "/oak/stanford/groups/akundaje/cmyun/chromatlas/reference/hg38.chrom.sizes",
+    "input_bam_file": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC/ENCSR382LBS/preprocessing/bigWigs/ENCSR382LBS.bigWig",
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "output_dir": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_0/",
+    "data_type": "ATAC",
+    "peaks": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_0/auxiliary/filtered.peaks.bed",
+    "nonpeaks": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_0/auxiliary/filtered.nonpeaks.bed",
+    "chr_fold_path": "/oak/stanford/groups/akundaje/cmyun/chromatlas/reference/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/cmyun/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/cmyun/chromatlas/bias_model/ENCFF881REW/fold_0/model.bias.fold_0.ENCSR868FGK.h5",
+    "negative_sampling_ratio": 0.1,
+    "filters": 512,
+    "n_dilation_layers": 8,
+    "max_jitter": 500,
+    "batch_size": 64,
+    "output_prefix": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_0/models/chrombpnet",
+    "bigwig": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_0/auxiliary/data_unstranded.bw",
+    "plus_shift": null,
+    "minus_shift": null,
+    "chr": "chr8",
+    "pwm_width": 24,
+    "params": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_0/logs/chrombpnet_model_params.tsv"
+}

fold_0/logs.models.fold_0.ENCSR382LBS/logfile.modelling.fold_0.ENCSR382LBS.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	4695.0
+trainings_pts_post_thresh	219648

fold_0/logs.models.fold_0.ENCSR382LBS/logfile.modelling.fold_0.ENCSR382LBS.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	26.9
+filters	512
+n_dil_layers	8
+bias_model_path	/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_0/models/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/oak/stanford/groups/akundaje/cmyun/chromatlas/reference/splits/fold_0.json
+negative_sampling_ratio	0.1

fold_0/logs.models.fold_0.ENCSR382LBS/logfile.modelling.fold_0.ENCSR382LBS.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
+0,1.4733163118362427,484.3629150390625,523.9953002929688,0.6838204264640808,491.59600830078125,509.9907531738281
+1,0.6778733730316162,459.2705078125,477.5050048828125,0.6291505694389343,480.8135681152344,497.7378845214844
+2,0.6236482262611389,450.498046875,467.2747802734375,0.6026655435562134,476.17169189453125,492.38372802734375
+3,0.5841235518455505,445.37066650390625,461.0849914550781,0.5735547542572021,474.3133239746094,489.7421569824219
+4,0.5495159029960632,441.4925537109375,456.2743225097656,0.5279672145843506,474.051025390625,488.2532958984375
+5,0.5232411623001099,438.1075744628906,452.1827697753906,0.5392002463340759,473.0723571777344,487.5770263671875
+6,0.5053008794784546,435.2665100097656,448.8586730957031,0.5458378195762634,474.8976745605469,489.580810546875
+7,0.48464104533195496,433.1380615234375,446.1750183105469,0.5121633410453796,473.1953125,486.9727478027344
+8,0.4639776349067688,431.90594482421875,444.38653564453125,0.6268713474273682,473.5855407714844,490.44854736328125
+9,0.44746193289756775,430.7087097167969,442.74517822265625,0.6235029697418213,473.5097961425781,490.28204345703125
+10,0.43012604117393494,429.97393798828125,441.5440368652344,0.5117930769920349,474.233154296875,488.0004577636719
+11,0.41823604702949524,429.1271667480469,440.3776550292969,0.5172789096832275,474.2858581542969,488.2005310058594
+12,0.4035889208316803,428.2107238769531,439.0674743652344,0.5589811205863953,473.9753723144531,489.0122375488281

fold_0/model.bias_scaled.fold_0.ENCSR382LBS.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b910e773052bbc4e2cf4d8e1cc5803afeb569ca4068cd4fadbbd80d201d1fcc8
+size 2691928

fold_0/model.bias_scaled.fold_0.ENCSR382LBS.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aefb47c5ff7d2f72303b80751d0e095a92be154cd63a4dfe0d1e7b99fb9560f1
+size 1198080

fold_0/model.chrombpnet.fold_0.ENCSR382LBS.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5e228b85adb89629c9a45104de0a01053b337aee82589eb07724cc24b29b5ba
+size 77538864

fold_0/model.chrombpnet.fold_0.ENCSR382LBS.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ed17496c91aec0fd42d2b9d33652290d1965315acb8a06eb009b1340393f4ff
+size 27535360

fold_0/model.chrombpnet_nobias.fold_0.ENCSR382LBS.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbc9f690bda7f9e414499dffdfefd33a5f5f8596fd70c5f6c0796783030deeb2
+size 25582648

fold_0/model.chrombpnet_nobias.fold_0.ENCSR382LBS.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:819045666c4bdb2d320d2f5852f8c4fe0518e1ca730f07cc92e2e271e0f97295
+size 26060800

fold_1/logs.models.fold_1.ENCSR382LBS/logfile.modelling.fold_1.ENCSR382LBS.args.json

@@ -0,0 +1,23 @@
+{
+    "genome": "/scratch/groups/akundaje/anusri/chromatin_atlas/reference/hg38.genome.fa",
+    "bigwig": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR382LBS//preprocessing/bigWigs/ENCSR382LBS.bigWig",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR382LBS//chrombpnet_model_feb15_fold_1//filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR382LBS//chrombpnet_model_feb15_fold_1//filtered.nonpeaks.bed",
+    "output_prefix": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR382LBS//chrombpnet_model_feb15_fold_1//chrombpnet",
+    "chr_fold_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/splits/fold_1.json",
+    "trackables": [
+        "logcount_predictions_loss",
+        "loss",
+        "logits_profile_predictions_loss",
+        "val_logcount_predictions_loss",
+        "val_loss",
+        "val_logits_profile_predictions_loss"
+    ],
+    "epochs": 50,
+    "early_stop": 5,
+    "batch_size": 64,
+    "learning_rate": 0.001,
+    "params": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR382LBS//chrombpnet_model_feb15_fold_1//chrombpnet_model_params.tsv",
+    "seed": 1234,
+    "architecture_from_file": "/home/users/anusri/chromatin-atlas-anvil/sherlock/chrombpnet/src/training/models/chrombpnet_with_bias_model.py"
+}

fold_1/logs.models.fold_1.ENCSR382LBS/logfile.modelling.fold_1.ENCSR382LBS.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	12.0
+counts_sum_max_thresh	2312.0
+trainings_pts_post_thresh	220874

fold_1/logs.models.fold_1.ENCSR382LBS/logfile.modelling.fold_1.ENCSR382LBS.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	27.4
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR382LBS//chrombpnet_model_feb15_fold_1/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/scratch/groups/akundaje/anusri/chromatin_atlas/splits/fold_1.json
+negative_sampling_ratio	0.1

fold_1/logs.models.fold_1.ENCSR382LBS/logfile.modelling.fold_1.ENCSR382LBS.epoch_loss.csv

@@ -0,0 +1,16 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,1.756470799446106,469.1395568847656,517.26708984375,0.5958902835845947,490.9732666015625,507.3008117675781
+1,0.6181320548057556,447.7062072753906,464.64324951171875,0.5518338680267334,479.82208251953125,494.9425048828125
+2,0.574364423751831,439.5935974121094,455.3310241699219,0.5077171921730042,474.2843933105469,488.1956787109375
+3,0.5369452834129333,435.33465576171875,450.0469055175781,0.4793228805065155,471.1492919921875,484.2828063964844
+4,0.5098628997802734,431.35675048828125,445.3262634277344,0.4698638617992401,470.22265625,483.09722900390625
+5,0.48616528511047363,428.7718505859375,442.0921630859375,0.4955636262893677,470.0470275878906,483.6253356933594
+6,0.46770593523979187,426.8321228027344,439.64654541015625,0.5320664644241333,470.2037658691406,484.78228759765625
+7,0.44840195775032043,425.4512939453125,437.7362365722656,0.5139513611793518,469.8209228515625,483.90325927734375
+8,0.3941604793071747,419.7762145996094,430.5761413574219,0.47806209325790405,468.4112854003906,481.51025390625
+9,0.36615362763404846,417.54913330078125,427.5812072753906,0.4454006552696228,468.9013977050781,481.1053161621094
+10,0.3410991430282593,415.59918212890625,424.945068359375,0.46153077483177185,470.50164794921875,483.1477355957031
+11,0.3198226988315582,414.25103759765625,423.0144958496094,0.48391789197921753,470.7492370605469,484.0089111328125
+12,0.30004245042800903,412.98504638671875,421.2066650390625,0.4810223877429962,471.295166015625,484.4754333496094
+13,0.26368600130081177,410.4413146972656,417.6668701171875,0.49339979887008667,471.5414123535156,485.0603332519531
+14,0.24990810453891754,409.16473388671875,416.01214599609375,0.4804895520210266,472.5047912597656,485.6702575683594

fold_1/model.bias_scaled.fold_1.ENCSR382LBS.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c4c5a76039402cfb1a899bcc20f6cc5f9abbd58721aa75989ef8ef500e1dc8b
+size 2688440

fold_1/model.bias_scaled.fold_1.ENCSR382LBS.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b947b8a0144df26db3c8899e598fc6bc0895ac4ddb2b7167081d064403f70fe
+size 1198080

fold_1/model.chrombpnet.fold_1.ENCSR382LBS.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c0a4d1098a19eed5b19a00b653eb214a509209d034883fe86af122c2139d570
+size 26448040

fold_1/model.chrombpnet.fold_1.ENCSR382LBS.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff1f35b3d9cf461909d55925825f07633a09b0956e876eaeb17b2be8e556d12d
+size 27525120

fold_1/model.chrombpnet_nobias.fold_1.ENCSR382LBS.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:777105ddaf114bf04b34f30a2b8dadff2a80616295b826e4da3b46df9585aa00
+size 25583536

fold_1/model.chrombpnet_nobias.fold_1.ENCSR382LBS.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c9ffed0de57ab3c1c19477eb9e687d9aeb4e98431f0ac84764139de15e0634e
+size 26060800

fold_2/logs.models.fold_2.ENCSR382LBS/logfile.modelling.fold_2.ENCSR382LBS.args.json

@@ -0,0 +1,50 @@
+{
+    "cmd": "pipeline",
+    "genome": "/oak/stanford/groups/akundaje/cmyun/chromatlas/reference/hg38.fa",
+    "chrom_sizes": "/oak/stanford/groups/akundaje/cmyun/chromatlas/reference/hg38.chrom.sizes",
+    "input_bam_file": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC/ENCSR382LBS/preprocessing/bigWigs/ENCSR382LBS.bigWig",
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "output_dir": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_2/",
+    "data_type": "ATAC",
+    "peaks": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_2/auxiliary/filtered.peaks.bed",
+    "nonpeaks": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_2/auxiliary/filtered.nonpeaks.bed",
+    "chr_fold_path": "/oak/stanford/groups/akundaje/cmyun/chromatlas/reference/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/cmyun/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/cmyun/chromatlas/bias_model/ENCFF881REW/fold_2/model.bias.fold_2.ENCSR868FGK.h5",
+    "negative_sampling_ratio": 0.1,
+    "filters": 512,
+    "n_dilation_layers": 8,
+    "max_jitter": 500,
+    "batch_size": 64,
+    "output_prefix": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_2/models/chrombpnet",
+    "bigwig": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_2/auxiliary/data_unstranded.bw",
+    "plus_shift": null,
+    "minus_shift": null,
+    "chr": "chr22",
+    "pwm_width": 24,
+    "params": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_2/logs/chrombpnet_model_params.tsv"
+}

fold_2/logs.models.fold_2.ENCSR382LBS/logfile.modelling.fold_2.ENCSR382LBS.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	5473.42
+trainings_pts_post_thresh	231155

fold_2/logs.models.fold_2.ENCSR382LBS/logfile.modelling.fold_2.ENCSR382LBS.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	27.4
+filters	512
+n_dil_layers	8
+bias_model_path	/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_2/models/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/oak/stanford/groups/akundaje/cmyun/chromatlas/reference/splits/fold_2.json
+negative_sampling_ratio	0.1

fold_2/logs.models.fold_2.ENCSR382LBS/logfile.modelling.fold_2.ENCSR382LBS.epoch_loss.csv

@@ -0,0 +1,17 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,4.659759998321533,488.33917236328125,616.0155029296875,0.8251886367797852,478.8931884765625,501.50335693359375
+1,0.7134513258934021,462.20867919921875,481.7569274902344,0.6486870646476746,466.2465515136719,484.0206298828125
+2,0.6636355519294739,453.0063781738281,471.19085693359375,0.6190513372421265,459.75848388671875,476.72052001953125
+3,0.6157399415969849,447.5379638671875,464.40948486328125,0.5700570344924927,458.396484375,474.0160217285156
+4,0.5779258012771606,443.34326171875,459.1779479980469,0.5914759635925293,454.30816650390625,470.51434326171875
+5,0.5479176640510559,439.66961669921875,454.6831359863281,0.5335731506347656,452.5655517578125,467.18560791015625
+6,0.5264312028884888,437.2375183105469,451.66241455078125,0.5416908860206604,452.7561950683594,467.59832763671875
+7,0.4982950687408447,434.8521728515625,448.5051574707031,0.6243636012077332,452.57745361328125,469.6848449707031
+8,0.48424312472343445,433.00732421875,446.2749328613281,0.6609243154525757,453.06103515625,471.17022705078125
+9,0.4623367190361023,431.58331298828125,444.2509460449219,0.5416407585144043,453.04022216796875,467.881103515625
+10,0.44424256682395935,430.7735900878906,442.9458312988281,0.5149888396263123,452.504638671875,466.6151428222656
+11,0.4265460968017578,429.4507141113281,441.1378479003906,0.5559778809547424,453.4779357910156,468.7119445800781
+12,0.4116733968257904,428.84063720703125,440.11956787109375,0.5068443417549133,453.3831481933594,467.2707824707031
+13,0.396022230386734,427.95294189453125,438.8040771484375,0.7433035969734192,453.98126220703125,474.3476867675781
+14,0.38552597165107727,427.5331115722656,438.09674072265625,0.5065503716468811,453.836669921875,467.7159729003906
+15,0.3726058006286621,427.3100280761719,437.52008056640625,0.5147143006324768,453.7032165527344,467.8064270019531

fold_2/model.bias_scaled.fold_2.ENCSR382LBS.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a63574502bf5527548d416f60fc652327707662b04f693444534b16d42b6dc64
+size 2688440

fold_2/model.bias_scaled.fold_2.ENCSR382LBS.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a699497a6ef981792eb4207e0a1f1550e6d60d0abcaf243962e8fa02d998d65
+size 1198080

fold_2/model.chrombpnet.fold_2.ENCSR382LBS.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e49258854d020c532aafdf31ee36e1432668632b374628bc68c41018e08d8a6
+size 77538808

fold_2/model.chrombpnet.fold_2.ENCSR382LBS.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e28d2943b45321df15359a8e85a30b03d9f68c6eef5d3401af5130adf2be0e4
+size 27525120

fold_2/model.chrombpnet_nobias.fold_2.ENCSR382LBS.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85e3c376c2b5dec19a6ede8bc4dc1b07a7375f882dcbf264183cb440fc3d5135
+size 25582648

fold_2/model.chrombpnet_nobias.fold_2.ENCSR382LBS.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3f0ef6ed4b5a408ccb3d39525f6fe48cc6ac7018dab70b2b2ba89a768c20315
+size 26060800

fold_3/logs.models.fold_3.ENCSR382LBS/logfile.modelling.fold_3.ENCSR382LBS.args.json

@@ -0,0 +1,23 @@
+{
+    "genome": "/scratch/groups/akundaje/anusri/chromatin_atlas/reference/hg38.genome.fa",
+    "bigwig": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR382LBS//preprocessing/bigWigs/ENCSR382LBS.bigWig",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR382LBS//chrombpnet_model_feb15_fold_3//filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR382LBS//chrombpnet_model_feb15_fold_3//filtered.nonpeaks.bed",
+    "output_prefix": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR382LBS//chrombpnet_model_feb15_fold_3//chrombpnet",
+    "chr_fold_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/splits/fold_3.json",
+    "trackables": [
+        "logcount_predictions_loss",
+        "loss",
+        "logits_profile_predictions_loss",
+        "val_logcount_predictions_loss",
+        "val_loss",
+        "val_logits_profile_predictions_loss"
+    ],
+    "epochs": 50,
+    "early_stop": 5,
+    "batch_size": 64,
+    "learning_rate": 0.001,
+    "params": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR382LBS//chrombpnet_model_feb15_fold_3//chrombpnet_model_params.tsv",
+    "seed": 1234,
+    "architecture_from_file": "/home/users/anusri/chromatin-atlas-anvil/sherlock/chrombpnet/src/training/models/chrombpnet_with_bias_model.py"
+}

fold_3/logs.models.fold_3.ENCSR382LBS/logfile.modelling.fold_3.ENCSR382LBS.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	12.0
+counts_sum_max_thresh	2286.33
+trainings_pts_post_thresh	222389

fold_3/logs.models.fold_3.ENCSR382LBS/logfile.modelling.fold_3.ENCSR382LBS.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	27.3
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR382LBS//chrombpnet_model_feb15_fold_3/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/scratch/groups/akundaje/anusri/chromatin_atlas/splits/fold_3.json
+negative_sampling_ratio	0.1

fold_3/logs.models.fold_3.ENCSR382LBS/logfile.modelling.fold_3.ENCSR382LBS.epoch_loss.csv

@@ -0,0 +1,17 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,1.5700547695159912,465.81683349609375,508.6795349121094,0.6133402585983276,466.6061706542969,483.3505859375
+1,0.5846941471099854,442.80145263671875,458.7623596191406,0.5507238507270813,457.3331298828125,472.3677062988281
+2,0.537192702293396,436.122314453125,450.7872619628906,0.6232540011405945,455.56121826171875,472.5759582519531
+3,0.5074432492256165,432.20806884765625,446.0600280761719,0.6158677935600281,453.65545654296875,470.4686584472656
+4,0.4870470464229584,429.1316223144531,442.42803955078125,0.5027739405632019,453.3614501953125,467.0873107910156
+5,0.46535524725914,426.9934997558594,439.6978759765625,0.5442349314689636,454.8256530761719,469.6833801269531
+6,0.4442325830459595,425.05157470703125,437.1791687011719,0.490639865398407,453.36224365234375,466.7567138671875
+7,0.4244338572025299,424.01104736328125,435.5984191894531,0.6178130507469177,454.04638671875,470.9126892089844
+8,0.40919411182403564,422.77960205078125,433.950927734375,0.4949021339416504,453.7874450683594,467.29852294921875
+9,0.3921850025653839,421.9778747558594,432.68524169921875,0.5194376707077026,455.0655517578125,469.2458190917969
+10,0.3364763557910919,416.8149719238281,426.00054931640625,0.5105211734771729,452.4307861328125,466.36810302734375
+11,0.3067052662372589,414.4833679199219,422.85675048828125,0.4980472922325134,453.1356506347656,466.7323303222656
+12,0.28738951683044434,413.1243591308594,420.969970703125,0.5124434232711792,453.93426513671875,467.9239196777344
+13,0.2692870497703552,411.9383850097656,419.29022216796875,0.5708307027816772,454.88519287109375,470.4686584472656
+14,0.23891191184520721,409.3413391113281,415.86419677734375,0.520790696144104,455.1144104003906,469.33148193359375
+15,0.22443604469299316,408.2229919433594,414.3501281738281,0.5380027294158936,455.1368103027344,469.8246154785156

fold_3/model.bias_scaled.fold_3.ENCSR382LBS.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:725c452aa2e2ab457c80fa492084ab5810d4fefe0beca5349a0d5c1efe3ca7ad
+size 2688440

fold_3/model.bias_scaled.fold_3.ENCSR382LBS.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9920137fc5be348f59bee63c317814583f1a0ecc8e08323035ae4de3843beb6b
+size 1198080

fold_3/model.chrombpnet.fold_3.ENCSR382LBS.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1cc7d20d9d3aa14b3da56f74ae5f3f9a85298645dd137177b00799ae5227e007
+size 26448040

fold_3/model.chrombpnet.fold_3.ENCSR382LBS.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e16fcbd202dcfacfe1a2858498d12bfdaef1c5b8e3c55f37635242c4ae948082
+size 27525120

fold_3/model.chrombpnet_nobias.fold_3.ENCSR382LBS.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04a66473d5ae8cde03bfe73878ccd032c8bcf5d7f7ca9e2671cddd0969b856da
+size 25583536

fold_3/model.chrombpnet_nobias.fold_3.ENCSR382LBS.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ccbb7aca896c21c194f728ccd0c111d3816f2f92bcfcdfce871b6c9d1b070d0
+size 26060800

fold_4/logs.models.fold_4.ENCSR382LBS/logfile.modelling.fold_4.ENCSR382LBS.args.json

@@ -0,0 +1,50 @@
+{
+    "cmd": "pipeline",
+    "genome": "/oak/stanford/groups/akundaje/cmyun/chromatlas/reference/hg38.fa",
+    "chrom_sizes": "/oak/stanford/groups/akundaje/cmyun/chromatlas/reference/hg38.chrom.sizes",
+    "input_bam_file": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC/ENCSR382LBS/preprocessing/bigWigs/ENCSR382LBS.bigWig",
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "output_dir": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_4/",
+    "data_type": "ATAC",
+    "peaks": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_4/auxiliary/filtered.peaks.bed",
+    "nonpeaks": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_4/auxiliary/filtered.nonpeaks.bed",
+    "chr_fold_path": "/oak/stanford/groups/akundaje/cmyun/chromatlas/reference/splits/fold_4.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/cmyun/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/cmyun/chromatlas/bias_model/ENCFF881REW/fold_4/model.bias.fold_4.ENCSR868FGK.h5",
+    "negative_sampling_ratio": 0.1,
+    "filters": 512,
+    "n_dilation_layers": 8,
+    "max_jitter": 500,
+    "batch_size": 64,
+    "output_prefix": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_4/models/chrombpnet",
+    "bigwig": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_4/auxiliary/data_unstranded.bw",
+    "plus_shift": null,
+    "minus_shift": null,
+    "chr": "chr10",
+    "pwm_width": 24,
+    "params": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_4/logs/chrombpnet_model_params.tsv"
+}

fold_4/logs.models.fold_4.ENCSR382LBS/logfile.modelling.fold_4.ENCSR382LBS.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	5490.26
+trainings_pts_post_thresh	238333

fold_4/logs.models.fold_4.ENCSR382LBS/logfile.modelling.fold_4.ENCSR382LBS.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	27.2
+filters	512
+n_dil_layers	8
+bias_model_path	/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR382LBS/chrombpnet_model_fold_4/models/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/oak/stanford/groups/akundaje/cmyun/chromatlas/reference/splits/fold_4.json
+negative_sampling_ratio	0.1

fold_4/logs.models.fold_4.ENCSR382LBS/logfile.modelling.fold_4.ENCSR382LBS.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
+0,1.5099844932556152,482.3714599609375,523.4434204101562,0.6693742275238037,463.63372802734375,481.84075927734375
+1,0.681018590927124,458.2076721191406,476.7317810058594,0.6076856255531311,455.8652038574219,472.39410400390625
+2,0.6217669248580933,450.6565856933594,467.56976318359375,0.5748499035835266,449.06390380859375,464.6999206542969
+3,0.5782055258750916,444.90753173828125,460.6343994140625,0.6027485132217407,447.6675109863281,464.0622863769531
+4,0.5338323712348938,441.4897766113281,456.01019287109375,0.51473468542099,445.6781005859375,459.6788330078125
+5,0.5077969431877136,438.48260498046875,452.2951354980469,0.5146594643592834,444.31280517578125,458.3114929199219
+6,0.4874672293663025,435.9473571777344,449.2066650390625,0.5191584825515747,444.2926330566406,458.4138488769531
+7,0.46564459800720215,434.14715576171875,446.8132629394531,0.5476580858230591,445.91265869140625,460.8091125488281
+8,0.4431893527507782,432.88116455078125,444.9365234375,0.5092480778694153,444.89593505859375,458.7474365234375
+9,0.43031764030456543,431.7569580078125,443.461669921875,0.5884107351303101,445.10626220703125,461.1109619140625
+10,0.41618481278419495,430.9211730957031,442.241455078125,0.49467775225639343,445.0790710449219,458.534423828125