Upload folder using huggingface_hub

README.md

@@ -0,0 +1,120 @@
+---
+license: mit
+library_name: chrombpnet
+tags:
+  - encode
+  - chrombpnet
+  - chromatin-accessibility
+  - ATAC
+  - breast
+  - 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 breast epithelium (ENCSR654UYP)
+- Model: ChromBPNet
+- Assay: ATAC-seq
+- Experiment: [ENCSR654UYP](https://www.encodeproject.org/experiments/ENCSR654UYP/)
+- Model annotation: [ENCSR930HKW](https://www.encodeproject.org/annotations/ENCSR930HKW/)
+- Biosample: breast epithelium (Full name: Homo sapiens breast epithelium tissue female adult (53 years))
+- Cell slim(s): None
+- Organ slim(s): epithelium,breast
+- Developmental slim(s): None
+- System slim(s): None
+- 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.ENCSR654UYP/logfile.modelling.fold_0.ENCSR654UYP.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/ENCSR654UYP/preprocessing/bigWigs/ENCSR654UYP.bigWig",
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "output_dir": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR654UYP/chrombpnet_model_fold_0/",
+    "data_type": "ATAC",
+    "peaks": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR654UYP/chrombpnet_model_fold_0/auxiliary/filtered.peaks.bed",
+    "nonpeaks": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR654UYP/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/ENCSR654UYP/chrombpnet_model_fold_0/models/chrombpnet",
+    "bigwig": "/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR654UYP/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/ENCSR654UYP/chrombpnet_model_fold_0/logs/chrombpnet_model_params.tsv"
+}

fold_0/logs.models.fold_0.ENCSR654UYP/logfile.modelling.fold_0.ENCSR654UYP.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	2060.56
+trainings_pts_post_thresh	107499

fold_0/logs.models.fold_0.ENCSR654UYP/logfile.modelling.fold_0.ENCSR654UYP.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	11.8
+filters	512
+n_dil_layers	8
+bias_model_path	/oak/stanford/groups/akundaje/cmyun/chromatlas/ATAC/ENCSR654UYP/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.ENCSR654UYP/logfile.modelling.fold_0.ENCSR654UYP.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
+0,1.5694915056228638,305.29803466796875,323.81781005859375,0.457076758146286,305.2379455566406,310.6315002441406
+1,0.333342045545578,296.39666748046875,300.32989501953125,0.271537721157074,302.2377624511719,305.4419250488281
+2,0.3042293190956116,292.4081115722656,295.9980773925781,0.28402015566825867,299.5161437988281,302.86749267578125
+3,0.2915411591529846,290.19769287109375,293.6377868652344,0.2517031729221344,298.3533630371094,301.3233337402344
+4,0.27062711119651794,288.7224426269531,291.9159851074219,0.25282368063926697,298.05621337890625,301.0395202636719
+5,0.2553940415382385,286.81561279296875,289.8291931152344,0.23456823825836182,297.9905090332031,300.75830078125
+6,0.24839510023593903,285.4797668457031,288.4110107421875,0.26060330867767334,297.8963317871094,300.9715270996094
+7,0.24060165882110596,283.93463134765625,286.77374267578125,0.2584429383277893,298.6471862792969,301.6967468261719
+8,0.23076094686985016,282.31689453125,285.0397033691406,0.2288832813501358,298.4077453613281,301.1085510253906
+9,0.22127456963062286,281.18402099609375,283.79522705078125,0.22722671926021576,299.6098327636719,302.29095458984375
+10,0.21205507218837738,280.25244140625,282.75445556640625,0.23251190781593323,299.2931213378906,302.0367736816406

fold_0/model.bias_scaled.fold_0.ENCSR654UYP.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a108de6684d4d08bb29cc4735c5d303827063a5d63468213a10ee57c2f902831
+size 2691928

fold_0/model.bias_scaled.fold_0.ENCSR654UYP.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:993662c7e6d52fc48afb29a5a57c7ec7227b8604ed987db9e136d8b8a52c105b
+size 1198080

fold_0/model.chrombpnet.fold_0.ENCSR654UYP.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd84beb476e7bde4459abf5c4203c26d387cf1d5518af255ecbab7b41399e257
+size 77538864

fold_0/model.chrombpnet.fold_0.ENCSR654UYP.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f28a40b1f430fb59039eaf77abde7a8108dce2ccfb5fd5e12a39c966422f10f8
+size 27535360

fold_0/model.chrombpnet_nobias.fold_0.ENCSR654UYP.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b75003c9d537e2899705760c66c710d7b29f2a475d141a88d6391eef8645d80
+size 25582648

fold_0/model.chrombpnet_nobias.fold_0.ENCSR654UYP.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c37b875dda5ba1528e7ebebec6bc327ca2df428a4c8f9d3df4eb1e524743b36e
+size 26060800

fold_1/logs.models.fold_1.ENCSR654UYP/logfile.modelling.fold_1.ENCSR654UYP.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//ENCSR654UYP//preprocessing/bigWigs/ENCSR654UYP.bigWig",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//chrombpnet_model_feb15_fold_1//filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//chrombpnet_model_feb15_fold_1//filtered.nonpeaks.bed",
+    "output_prefix": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//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/ENCSR654UYP//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.ENCSR654UYP/logfile.modelling.fold_1.ENCSR654UYP.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	18.0
+counts_sum_max_thresh	864.0
+trainings_pts_post_thresh	108150

fold_1/logs.models.fold_1.ENCSR654UYP/logfile.modelling.fold_1.ENCSR654UYP.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	11.9
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//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.ENCSR654UYP/logfile.modelling.fold_1.ENCSR654UYP.epoch_loss.csv

@@ -0,0 +1,18 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,3.693004608154297,294.5162658691406,338.46356201171875,0.3957681655883789,322.7968444824219,327.506591796875
+1,0.3078840672969818,287.3446350097656,291.0085144042969,0.3030795156955719,317.0390930175781,320.6457214355469
+2,0.2786576747894287,283.92327880859375,287.2390441894531,0.2910393476486206,313.7337951660156,317.1971740722656
+3,0.2661859393119812,281.64373779296875,284.8111572265625,0.2714170515537262,312.3023681640625,315.53204345703125
+4,0.24775515496730804,279.97528076171875,282.92352294921875,0.2609420120716095,311.6294250488281,314.7347412109375
+5,0.24303670227527618,279.19866943359375,282.0910339355469,0.27018630504608154,310.9561767578125,314.1712951660156
+6,0.23066207766532898,278.16168212890625,280.9068603515625,0.2777594029903412,310.1942443847656,313.4996643066406
+7,0.2235822081565857,276.8345031738281,279.4948425292969,0.2679443955421448,310.0820617675781,313.2706298828125
+8,0.2145320326089859,276.2677001953125,278.8206787109375,0.2785249352455139,310.41766357421875,313.73199462890625
+9,0.20297403633594513,275.3979187011719,277.8134765625,0.23519442975521088,310.7005920410156,313.4992980957031
+10,0.19645555317401886,274.1186828613281,276.45654296875,0.23729752004146576,310.7166748046875,313.54046630859375
+11,0.16896972060203552,271.0152893066406,273.026123046875,0.2326485812664032,310.3423767089844,313.11102294921875
+12,0.15828070044517517,268.8147277832031,270.6985168457031,0.249937042593956,311.23040771484375,314.2046203613281
+13,0.14899514615535736,267.2804260253906,269.0538330078125,0.23953215777873993,312.2145690917969,315.06494140625
+14,0.14220109581947327,265.672607421875,267.36492919921875,0.23769648373126984,313.3593444824219,316.18792724609375
+15,0.12769848108291626,263.3175048828125,264.8374938964844,0.2412533462047577,314.6407165527344,317.5116271972656
+16,0.12326730042695999,261.96160888671875,263.42816162109375,0.24734370410442352,315.6210632324219,318.564453125

fold_1/model.bias_scaled.fold_1.ENCSR654UYP.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:117a026ba8b9eccd0ec95f1ae8f7f877cc0afde5e22916e786e1093fc7b47dd5
+size 2688440

fold_1/model.bias_scaled.fold_1.ENCSR654UYP.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fec8215a19b9b0e6106acf8179549e89102d0dee705b7d68a19cafafbca54de0
+size 1198080

fold_1/model.chrombpnet.fold_1.ENCSR654UYP.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:517990bb708530805085356164846da2cdb1a49ce26b87412be71a71c1b58c85
+size 26448040

fold_1/model.chrombpnet.fold_1.ENCSR654UYP.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15c0a89fbbf0bcccd5253d96bdc3627b190d00baabed1043e9c2fa6015710cf9
+size 27525120

fold_1/model.chrombpnet_nobias.fold_1.ENCSR654UYP.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8eb11e8bd4bc2f80e54749f8f1ade8e0e6fa3b8252ec6c00a428a42b0c2b116
+size 25583536

fold_1/model.chrombpnet_nobias.fold_1.ENCSR654UYP.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d59085e1076dd298d3b9d04492fd02412da60e58b50fa5957956abb4e248613
+size 26060800

fold_2/logs.models.fold_2.ENCSR654UYP/logfile.modelling.fold_2.ENCSR654UYP.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//ENCSR654UYP//preprocessing/bigWigs/ENCSR654UYP.bigWig",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//chrombpnet_model_feb15_fold_2//filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//chrombpnet_model_feb15_fold_2//filtered.nonpeaks.bed",
+    "output_prefix": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//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": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//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.ENCSR654UYP/logfile.modelling.fold_2.ENCSR654UYP.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	20.0
+counts_sum_max_thresh	865.0
+trainings_pts_post_thresh	110473

fold_2/logs.models.fold_2.ENCSR654UYP/logfile.modelling.fold_2.ENCSR654UYP.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	11.9
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//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.ENCSR654UYP/logfile.modelling.fold_2.ENCSR654UYP.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
+0,1.3398481607437134,293.5611572265625,309.50531005859375,0.32279375195503235,304.7766418457031,308.6179504394531
+1,0.27793624997138977,285.65673828125,288.9644470214844,0.30185550451278687,303.283935546875,306.8761901855469
+2,0.25740861892700195,282.34393310546875,285.407470703125,0.2973523437976837,299.0659484863281,302.6045227050781
+3,0.24492819607257843,280.8072814941406,283.72222900390625,0.29504746198654175,298.393310546875,301.90435791015625
+4,0.23122785985469818,278.8569641113281,281.60870361328125,0.2924182713031769,297.8472595214844,301.3270568847656
+5,0.2197544425725937,277.83026123046875,280.4454040527344,0.2507653832435608,297.3988037109375,300.38287353515625
+6,0.21059702336788177,276.4337463378906,278.9397888183594,0.2641080617904663,297.73828125,300.8812561035156
+7,0.20283611118793488,275.51312255859375,277.9266052246094,0.23925262689590454,297.46368408203125,300.31072998046875
+8,0.19457188248634338,274.1510009765625,276.4661560058594,0.231708362698555,297.79107666015625,300.54833984375
+9,0.18265840411186218,273.07366943359375,275.24713134765625,0.23447389900684357,298.1920166015625,300.9822692871094
+10,0.17736409604549408,272.47332763671875,274.5835876464844,0.22620047628879547,298.4728088378906,301.1645202636719
+11,0.15376856923103333,269.0060119628906,270.8357238769531,0.22730402648448944,298.8304748535156,301.535400390625
+12,0.14355036616325378,267.05621337890625,268.7646179199219,0.22741229832172394,300.0188293457031,302.7249450683594

fold_2/model.bias_scaled.fold_2.ENCSR654UYP.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d3a80bab81e71058e0600e91b26f205036d60d39bb896618e7251d7835fb846
+size 2688440

fold_2/model.bias_scaled.fold_2.ENCSR654UYP.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d72924a271474b7e92b365c8f533d8f0f50682a8a298eda45996b3081c939cf
+size 1198080

fold_2/model.chrombpnet.fold_2.ENCSR654UYP.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c3c22b81de478b32059fa5eaa655aceeabe815f68a2e4b67900136df28c2f8b
+size 26448040

fold_2/model.chrombpnet.fold_2.ENCSR654UYP.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe55db14f20a192ccf0447c3001bcf020274ba9e7e27d2d522ccc717b0072a72
+size 27525120

fold_2/model.chrombpnet_nobias.fold_2.ENCSR654UYP.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2765540a94ec4aa379079e9588075d3949b638411d90d9f9d0e664cf1803a39b
+size 25583536

fold_2/model.chrombpnet_nobias.fold_2.ENCSR654UYP.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8846f6d5c514d81d0e11bc4717f2a00eda1fbdef2afe90db324bedee862d26e
+size 26060800

fold_3/logs.models.fold_3.ENCSR654UYP/logfile.modelling.fold_3.ENCSR654UYP.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//ENCSR654UYP//preprocessing/bigWigs/ENCSR654UYP.bigWig",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//chrombpnet_model_feb15_fold_3//filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//chrombpnet_model_feb15_fold_3//filtered.nonpeaks.bed",
+    "output_prefix": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//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/ENCSR654UYP//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.ENCSR654UYP/logfile.modelling.fold_3.ENCSR654UYP.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	18.0
+counts_sum_max_thresh	875.0
+trainings_pts_post_thresh	108941

fold_3/logs.models.fold_3.ENCSR654UYP/logfile.modelling.fold_3.ENCSR654UYP.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	11.9
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//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.ENCSR654UYP/logfile.modelling.fold_3.ENCSR654UYP.epoch_loss.csv

@@ -0,0 +1,11 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,1.4985591173171997,296.01165771484375,313.84490966796875,0.3204001188278198,298.48577880859375,302.298583984375
+1,0.29164576530456543,287.32000732421875,290.79052734375,0.27312901616096497,295.263427734375,298.5137023925781
+2,0.2608184516429901,284.645751953125,287.7496337890625,0.26514169573783875,292.4185485839844,295.5737609863281
+3,0.24084730446338654,282.2174377441406,285.083251953125,0.25306791067123413,292.1326599121094,295.14410400390625
+4,0.23158858716487885,281.1811218261719,283.9369812011719,0.226212739944458,291.27935791015625,293.97125244140625
+5,0.21683481335639954,279.59393310546875,282.17474365234375,0.22982321679592133,292.5472106933594,295.2821350097656
+6,0.21149401366710663,278.1412658691406,280.658203125,0.25742676854133606,291.6775817871094,294.7409362792969
+7,0.2029612511396408,277.1834716796875,279.5991516113281,0.21318019926548004,291.5671081542969,294.1040344238281
+8,0.17482981085777283,273.8926086425781,275.97314453125,0.2592026889324188,291.3246154785156,294.4091491699219
+9,0.1655505746603012,271.7641296386719,273.7343444824219,0.22055810689926147,291.9517517089844,294.5764465332031

fold_3/model.bias_scaled.fold_3.ENCSR654UYP.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a170496b00d4dbd5ba65bc001146d465cb3c12249517db1f7d760cfd039bdb1
+size 2688440

fold_3/model.bias_scaled.fold_3.ENCSR654UYP.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:002628ff6342fd26d7db1772535ced064ed155391b9380754710e482ed716f4d
+size 1198080

fold_3/model.chrombpnet.fold_3.ENCSR654UYP.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52cb2894b6854d48c84f1a842d215620caaa775e947e7d6380fdcd365bf9adaa
+size 26448040

fold_3/model.chrombpnet.fold_3.ENCSR654UYP.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2dbcbbf5ae953d5a131ba28bafb27e4f96a286863def78263019d0259de3bacb
+size 27525120

fold_3/model.chrombpnet_nobias.fold_3.ENCSR654UYP.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b89a4b272968b0b03d5699555c78f3af2d771fa24987c7b1a765ae8f03e4227
+size 25583536

fold_3/model.chrombpnet_nobias.fold_3.ENCSR654UYP.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:913050be3c152a25fdc2589a21c2747812ad30dc4b3c7bdcf7fadf1b512f9233
+size 26060800

fold_4/logs.models.fold_4.ENCSR654UYP/logfile.modelling.fold_4.ENCSR654UYP.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//ENCSR654UYP//preprocessing/bigWigs/ENCSR654UYP.bigWig",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//chrombpnet_model_feb15_fold_4//filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//chrombpnet_model_feb15_fold_4//filtered.nonpeaks.bed",
+    "output_prefix": "/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//chrombpnet_model_feb15_fold_4//chrombpnet",
+    "chr_fold_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/splits/fold_4.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/ENCSR654UYP//chrombpnet_model_feb15_fold_4//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_4/logs.models.fold_4.ENCSR654UYP/logfile.modelling.fold_4.ENCSR654UYP.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	19.0
+counts_sum_max_thresh	858.0
+trainings_pts_post_thresh	109111

fold_4/logs.models.fold_4.ENCSR654UYP/logfile.modelling.fold_4.ENCSR654UYP.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	11.7
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/ATAC/ENCSR654UYP//chrombpnet_model_feb15_fold_4/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/scratch/groups/akundaje/anusri/chromatin_atlas/splits/fold_4.json
+negative_sampling_ratio	0.1

fold_4/logs.models.fold_4.ENCSR654UYP/logfile.modelling.fold_4.ENCSR654UYP.epoch_loss.csv

@@ -0,0 +1,13 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,1.00563645362854,290.876953125,302.64263916015625,0.2691979706287384,291.8873291015625,295.0369567871094
+1,0.2688491940498352,283.6439514160156,286.78924560546875,0.2326972633600235,289.1217041015625,291.8441467285156
+2,0.24413828551769257,280.7650146484375,283.6211242675781,0.219515860080719,288.2203674316406,290.788818359375
+3,0.22828476130962372,279.0767822265625,281.7475891113281,0.2360997349023819,288.3077392578125,291.0699768066406
+4,0.21585673093795776,277.2822265625,279.8079833984375,0.2345791906118393,287.6770324707031,290.4216003417969
+5,0.20587581396102905,275.8279724121094,278.2365417480469,0.21406207978725433,287.8246765136719,290.3292236328125
+6,0.19925233721733093,274.71392822265625,277.0453186035156,0.19725143909454346,287.8294372558594,290.13726806640625
+7,0.18963655829429626,273.1954040527344,275.4140319824219,0.18603697419166565,288.3537292480469,290.5302429199219
+8,0.18443627655506134,272.6430969238281,274.8012390136719,0.18620873987674713,288.7852478027344,290.9638366699219
+9,0.17016500234603882,271.5146179199219,273.5057678222656,0.1953122466802597,289.18023681640625,291.4654541015625
+10,0.1487313061952591,268.2160339355469,269.9564208984375,0.19043667614459991,289.65789794921875,291.88592529296875
+11,0.1394205242395401,266.21026611328125,267.8416442871094,0.18757760524749756,290.65667724609375,292.85137939453125