Upload folder using huggingface_hub

README.md

@@ -0,0 +1,120 @@
+---
+license: mit
+library_name: chrombpnet
+tags:
+  - encode
+  - chrombpnet
+  - chromatin-accessibility
+  - ATAC
+  - lung
+  - 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 upper lobe of left lung (ENCSR895SEG)
+- Model: ChromBPNet
+- Assay: ATAC-seq
+- Experiment: [ENCSR895SEG](https://www.encodeproject.org/experiments/ENCSR895SEG/)
+- Model annotation: [ENCSR388FHS](https://www.encodeproject.org/annotations/ENCSR388FHS/)
+- Biosample: upper lobe of left lung (Full name: Homo sapiens upper lobe of left lung tissue male adult (60 years))
+- Cell slim(s): None
+- Organ slim(s): lung
+- Developmental slim(s): endoderm
+- System slim(s): respiratory-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.ENCSR895SEG/logfile.modelling.fold_0.ENCSR895SEG.args.json

@@ -0,0 +1,23 @@
+{
+    "genome": "/scratch/groups/akundaje/anusri/chromatin_atlas/reference/hg38.genome.fa",
+    "bigwig": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR895SEG//preprocessing/bigWigs/ENCSR895SEG.bigWig",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR895SEG//chrombpnet_model_feb15//filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR895SEG//chrombpnet_model_feb15//filtered.nonpeaks.bed",
+    "output_prefix": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR895SEG//chrombpnet_model_feb15//chrombpnet",
+    "chr_fold_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/splits/fold_0.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/ENCSR895SEG//chrombpnet_model_feb15//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_0/logs.models.fold_0.ENCSR895SEG/logfile.modelling.fold_0.ENCSR895SEG.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC/ENCSR895SEG/chrombpnet_model_feb15/bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/fold_0/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR895SEG/logfile.modelling.fold_0.ENCSR895SEG.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	78.0
+counts_sum_max_thresh	4212.84
+trainings_pts_post_thresh	188048

fold_0/logs.models.fold_0.ENCSR895SEG/logfile.modelling.fold_0.ENCSR895SEG.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC/ENCSR895SEG/chrombpnet_model_feb15/chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/fold_0/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR895SEG/logfile.modelling.fold_0.ENCSR895SEG.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	41.6
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR895SEG//chrombpnet_model_feb15/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/scratch/groups/akundaje/anusri/chromatin_atlas/splits/fold_0.json
+negative_sampling_ratio	0.1

fold_0/logs.models.fold_0.ENCSR895SEG/logfile.modelling.fold_0.ENCSR895SEG.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC/ENCSR895SEG/chrombpnet_model_feb15/chrombpnet_wo_bias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/fold_0/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR895SEG/logfile.modelling.fold_0.ENCSR895SEG.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
+0,0.6125558018684387,668.6963500976562,694.1771240234375,0.32411476969718933,657.79248046875,671.2755737304688
+1,0.2781808078289032,643.7119140625,655.2838745117188,0.2450685203075409,645.4014892578125,655.596435546875
+2,0.2573201060295105,635.2619018554688,645.9671020507812,0.24143123626708984,645.2247314453125,655.2680053710938
+3,0.23881682753562927,629.0892944335938,639.0230102539062,0.23371806740760803,644.632568359375,654.35546875
+4,0.22654248774051666,625.4822998046875,634.9071044921875,0.23049093782901764,643.877685546875,653.4663696289062
+5,0.21487663686275482,621.173828125,630.1142578125,0.22041331231594086,642.9821166992188,652.1513061523438
+6,0.20684072375297546,618.7332763671875,627.33837890625,0.21868035197257996,643.438720703125,652.5357055664062
+7,0.1978352665901184,616.6475830078125,624.877197265625,0.2469264715909958,642.95556640625,653.2278442382812
+8,0.19076795876026154,615.2976684570312,623.2342529296875,0.22866863012313843,644.856689453125,654.3690795898438
+9,0.1626109927892685,607.591552734375,614.3572387695312,0.21389174461364746,640.0921630859375,648.9898071289062
+10,0.14883409440517426,602.1074829101562,608.2986450195312,0.23929007351398468,641.407958984375,651.3624267578125
+11,0.13876751065254211,599.53173828125,605.3053588867188,0.22077293694019318,641.55322265625,650.7371215820312
+12,0.12897461652755737,598.0314331054688,603.3978271484375,0.22498719394207,642.2083740234375,651.5677490234375
+13,0.11420223861932755,593.6248168945312,598.3767700195312,0.2254447042942047,642.10107421875,651.4790649414062
+14,0.10814183950424194,591.5584106445312,596.0565795898438,0.222579687833786,642.6240234375,651.8829956054688

fold_0/model.bias_scaled.fold_0.ENCSR895SEG.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf9906a547ea6b63f0a165e6d6df09c760dd013da94536e27dab82198478155a
+size 2688440

fold_0/model.bias_scaled.fold_0.ENCSR895SEG.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:433bcf1b45cd817dead7ae57ce45d9cde133a35fcfb80616e8ebbcedde6ac15e
+size 1198080

fold_0/model.chrombpnet.fold_0.ENCSR895SEG.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:648b747eb78b7549dcaf838b36fc811378f4ed37bf21aaef8403954ff001cfc4
+size 26448016

fold_0/model.chrombpnet.fold_0.ENCSR895SEG.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddf08c94e7d9361dac3ae00efb0c165141e146a0d2b9b1def51ee97af048506c
+size 27535360

fold_0/model.chrombpnet_nobias.fold_0.ENCSR895SEG.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:357bed9c1f1e141d1311099caac3a9a06228c1b2ce5c79f6e7980d56c43a4d00
+size 25583536

fold_0/model.chrombpnet_nobias.fold_0.ENCSR895SEG.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca3d06fc4bb872f83253fc52494df27918cb6218b8c45aee8c8a766b44b476af
+size 26060800

fold_1/logs.models.fold_1.ENCSR895SEG/logfile.modelling.fold_1.ENCSR895SEG.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//ENCSR895SEG//preprocessing/bigWigs/ENCSR895SEG.bigWig",
+    "peaks": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//chrombpnet_model_feb15_fold_1//filtered.peaks.bed",
+    "nonpeaks": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//chrombpnet_model_feb15_fold_1//filtered.nonpeaks.bed",
+    "output_prefix": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//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": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//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.ENCSR895SEG/logfile.modelling.fold_1.ENCSR895SEG.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/chrombpnet_model_feb15_fold_1/bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/fold_1/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR895SEG/logfile.modelling.fold_1.ENCSR895SEG.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	82.0
+counts_sum_max_thresh	4197.48
+trainings_pts_post_thresh	190849

fold_1/logs.models.fold_1.ENCSR895SEG/logfile.modelling.fold_1.ENCSR895SEG.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/chrombpnet_model_feb15_fold_1/chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/fold_1/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR895SEG/logfile.modelling.fold_1.ENCSR895SEG.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	41.8
+filters	512
+n_dil_layers	8
+bias_model_path	/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//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.ENCSR895SEG/logfile.modelling.fold_1.ENCSR895SEG.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/chrombpnet_model_feb15_fold_1/chrombpnet_wo_bias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/fold_1/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR895SEG/logfile.modelling.fold_1.ENCSR895SEG.epoch_loss.csv

@@ -0,0 +1,19 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,0.8946594595909119,648.1194458007812,685.515380859375,0.3197689354419708,683.8460083007812,697.2125854492188
+1,0.27303561568260193,621.1565551757812,632.5707397460938,0.30493229627609253,675.3538818359375,688.1000366210938
+2,0.25236958265304565,613.5504150390625,624.0993041992188,0.24634116888046265,671.002197265625,681.2991333007812
+3,0.23635227978229523,608.3541259765625,618.2340698242188,0.32624486088752747,669.5272827148438,683.1641845703125
+4,0.22311297059059143,604.856689453125,614.1830444335938,0.23073947429656982,668.9120483398438,678.5575561523438
+5,0.2086162567138672,601.8745727539062,610.5955810546875,0.22511205077171326,668.3655395507812,677.7752075195312
+6,0.1973600834608078,599.4886474609375,607.7392578125,0.2206076979637146,668.375,677.5972290039062
+7,0.18661539256572723,597.7807006835938,605.5812377929688,0.24326598644256592,668.0028686523438,678.1712646484375
+8,0.17548075318336487,596.2090454101562,603.5455932617188,0.21807442605495453,668.1353759765625,677.2514038085938
+9,0.16716137528419495,594.9169311523438,601.9046020507812,0.23420940339565277,668.27783203125,678.067626953125
+10,0.1602039635181427,593.6504516601562,600.3468627929688,0.21819941699504852,668.4833984375,677.6043090820312
+11,0.15347696840763092,592.6544799804688,599.0698852539062,0.2228468805551529,670.2225341796875,679.53759765625
+12,0.12505272030830383,586.6603393554688,591.887939453125,0.22249600291252136,665.8751220703125,675.1756591796875
+13,0.11353115737438202,583.8785400390625,588.6242065429688,0.21874794363975525,666.0745849609375,675.2182006835938
+14,0.10395536571741104,582.1734008789062,586.5184936523438,0.22450636327266693,666.7293090820312,676.1130981445312
+15,0.09824563562870026,580.3015747070312,584.408447265625,0.21754375100135803,667.47998046875,676.5734252929688
+16,0.08586762845516205,578.025634765625,581.6143798828125,0.22064639627933502,667.4346313476562,676.6575317382812
+17,0.08180751651525497,576.15576171875,579.5762939453125,0.23757824301719666,668.1072998046875,678.0382690429688

fold_1/model.bias_scaled.fold_1.ENCSR895SEG.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e76c55b426de4bcf0c7eff0fac90d2f25ab6e13413b4c378efba05f5f032ca00
+size 2688440

fold_1/model.bias_scaled.fold_1.ENCSR895SEG.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9520d3eff7724843f260967a502d091e9e5d6c87abd98a62eafaab024b432fbf
+size 1198080

fold_1/model.chrombpnet.fold_1.ENCSR895SEG.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:296fcd2e207f9fe0c361378a0dfbd67c68c5ad55a4c0d12434a513ac602c9216
+size 26447928

fold_1/model.chrombpnet.fold_1.ENCSR895SEG.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:206fb8200089d2e45c09dbfe93b5d806a3f04ba6b145d5f0165f4bced58651c8
+size 27525120

fold_1/model.chrombpnet_nobias.fold_1.ENCSR895SEG.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ec6b97b037983b0378c20d81cca49f7cd30bc404ac2eaed7221b6121a10651a
+size 25583536

fold_1/model.chrombpnet_nobias.fold_1.ENCSR895SEG.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54d030e3ac773fd6f8fbcd772f6b513a7bb53d3cea17b43a154e7f7b0ce98a6d
+size 26060800

fold_2/logs.models.fold_2.ENCSR895SEG/logfile.modelling.fold_2.ENCSR895SEG.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//ENCSR895SEG//preprocessing/bigWigs/ENCSR895SEG.bigWig",
+    "peaks": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//chrombpnet_model_feb15_fold_2//filtered.peaks.bed",
+    "nonpeaks": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//chrombpnet_model_feb15_fold_2//filtered.nonpeaks.bed",
+    "output_prefix": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//chrombpnet_model_feb15_fold_2//chrombpnet",
+    "chr_fold_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/splits/fold_2.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": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//chrombpnet_model_feb15_fold_2//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_2/logs.models.fold_2.ENCSR895SEG/logfile.modelling.fold_2.ENCSR895SEG.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/chrombpnet_model_feb15_fold_2/bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/fold_2/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR895SEG/logfile.modelling.fold_2.ENCSR895SEG.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	85.0
+counts_sum_max_thresh	4250.28
+trainings_pts_post_thresh	197341

fold_2/logs.models.fold_2.ENCSR895SEG/logfile.modelling.fold_2.ENCSR895SEG.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/chrombpnet_model_feb15_fold_2/chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/fold_2/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR895SEG/logfile.modelling.fold_2.ENCSR895SEG.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	41.6
+filters	512
+n_dil_layers	8
+bias_model_path	/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//chrombpnet_model_feb15_fold_2/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/scratch/groups/akundaje/anusri/chromatin_atlas/splits/fold_2.json
+negative_sampling_ratio	0.1

fold_2/logs.models.fold_2.ENCSR895SEG/logfile.modelling.fold_2.ENCSR895SEG.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/chrombpnet_model_feb15_fold_2/chrombpnet_wo_bias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/fold_2/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR895SEG/logfile.modelling.fold_2.ENCSR895SEG.epoch_loss.csv

@@ -0,0 +1,19 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,1.2680984735488892,650.5484619140625,703.3015747070312,0.2950281798839569,650.9700927734375,663.2428588867188
+1,0.27635955810546875,623.6727294921875,635.1687622070312,0.2544699013233185,643.3236694335938,653.9098510742188
+2,0.2510841190814972,615.5955200195312,626.0394287109375,0.23736286163330078,636.7761840820312,646.6506958007812
+3,0.23327568173408508,610.3460083007812,620.0504760742188,0.2299727201461792,635.5809936523438,645.1480102539062
+4,0.21875213086605072,606.9090576171875,616.0092163085938,0.24934586882591248,635.0389404296875,645.4118041992188
+5,0.20795048773288727,603.8472900390625,612.4994506835938,0.22216159105300903,634.8320922851562,644.073974609375
+6,0.1959427297115326,601.6845703125,609.8348388671875,0.22277596592903137,634.2680053710938,643.5355834960938
+7,0.1873362511396408,599.7525024414062,607.5457153320312,0.23029524087905884,634.9492797851562,644.5298461914062
+8,0.17780467867851257,597.8666381835938,605.2638549804688,0.21980305016040802,633.798828125,642.9424438476562
+9,0.16989301145076752,596.4295654296875,603.4966430664062,0.21701973676681519,634.7776489257812,643.805419921875
+10,0.16263853013515472,595.5094604492188,602.2743530273438,0.21430586278438568,635.135986328125,644.05078125
+11,0.1549457609653473,594.5521850585938,600.9982299804688,0.2351645976305008,635.2421264648438,645.025146484375
+12,0.12853772938251495,588.8073120117188,594.1541748046875,0.21612301468849182,632.3875732421875,641.3782958984375
+13,0.1153450459241867,585.5535888671875,590.3521728515625,0.2218872308731079,632.98876953125,642.2193603515625
+14,0.1064995676279068,584.2920532226562,588.7219848632812,0.22238034009933472,633.1862182617188,642.437255859375
+15,0.10045263916254044,582.6648559570312,586.8431396484375,0.220102921128273,633.3893432617188,642.544921875
+16,0.08808618783950806,580.2028198242188,583.8682250976562,0.2238416224718094,633.2936401367188,642.6055297851562
+17,0.08279217034578323,578.6312866210938,582.0756225585938,0.2237485945224762,633.83984375,643.147705078125

fold_2/model.bias_scaled.fold_2.ENCSR895SEG.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8ee2df5c79e4555dba940cd90bbf2674bc56a1871037a786951ce0d2037ff5e
+size 2688440

fold_2/model.bias_scaled.fold_2.ENCSR895SEG.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a9e0e40965bef885737b393e451b2bd6d8595f66e00368caa0d059aeaa4767f
+size 1198080

fold_2/model.chrombpnet.fold_2.ENCSR895SEG.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f2653686e0bd7d50d73f47a67bb46893fde560ff4e4f3401d1ba4e63c57a8e9
+size 26447928

fold_2/model.chrombpnet.fold_2.ENCSR895SEG.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed1ae4385e4c7c261a114c801ec24cbf9713376f95ac2943a028fd66d5aad44c
+size 27525120

fold_2/model.chrombpnet_nobias.fold_2.ENCSR895SEG.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72ca57b1f2ffc0d248547be2430a11afea9a4f488412b4f29383681b5067b88c
+size 25583536

fold_2/model.chrombpnet_nobias.fold_2.ENCSR895SEG.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:328e5d9b4401a90bdfe8508a15da774dea59d5ba7a400dc0b060abf9db0eae20
+size 26060800

fold_3/logs.models.fold_3.ENCSR895SEG/logfile.modelling.fold_3.ENCSR895SEG.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//ENCSR895SEG//preprocessing/bigWigs/ENCSR895SEG.bigWig",
+    "peaks": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//chrombpnet_model_feb15_fold_3//filtered.peaks.bed",
+    "nonpeaks": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//chrombpnet_model_feb15_fold_3//filtered.nonpeaks.bed",
+    "output_prefix": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//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": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//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.ENCSR895SEG/logfile.modelling.fold_3.ENCSR895SEG.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/chrombpnet_model_feb15_fold_3/bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/fold_3/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_3/logs.models.fold_3.ENCSR895SEG/logfile.modelling.fold_3.ENCSR895SEG.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	77.0
+counts_sum_max_thresh	4239.64
+trainings_pts_post_thresh	190886

fold_3/logs.models.fold_3.ENCSR895SEG/logfile.modelling.fold_3.ENCSR895SEG.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/chrombpnet_model_feb15_fold_3/chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/fold_3/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_3/logs.models.fold_3.ENCSR895SEG/logfile.modelling.fold_3.ENCSR895SEG.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	41.8
+filters	512
+n_dil_layers	8
+bias_model_path	/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/ATAC//ENCSR895SEG//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.ENCSR895SEG/logfile.modelling.fold_3.ENCSR895SEG.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/chrombpnet_model_feb15_fold_3/chrombpnet_wo_bias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR895SEG/fold_3/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py