Upload folder using huggingface_hub

README.md

@@ -0,0 +1,120 @@
+---
+license: mit
+library_name: chrombpnet
+tags:
+  - encode
+  - chrombpnet
+  - chromatin-accessibility
+  - DNASE
+  - gyrus
+  - hg38
+---
+# ENCODE ChromBPNet Atlas
+As part of the ENCODE 4 Project, we trained ChromBPNet models on 1,512 ENCODE DNAse-seq and ATAC-seq across 408 biosamples. Here, we provide all models for open-source use.
+
+For more information about the models, see:
+- Main ENCODE 4 Paper
+- [A unified lexicon of predictive DNA sequence motifs from ENCODE transcription factor binding and chromatin accessibility assays](https://doi.org/10.5281/zenodo.17123347) (Deshpande et al., Zenodo 2025)
+- [ChromBPNet: bias factorized, base-resolution deep learning models of chromatin accessibility reveal cis-regulatory sequence syntax, transcription factor footprints and regulatory variants](https://doi.org/10.1101/2024.12.25.630221) (Pampari et al., bioRxiv 2024)
+
+## ChromBPNet model: DNASE in posterior cingulate gyrus (ENCSR590EFT)
+- Model: ChromBPNet
+- Assay: DNASE-seq
+- Experiment: [ENCSR590EFT](https://www.encodeproject.org/experiments/ENCSR590EFT/)
+- Model annotation: [ENCSR122RRZ](https://www.encodeproject.org/annotations/ENCSR122RRZ/)
+- Biosample: posterior cingulate gyrus (Full name: Homo sapiens posterior cingulate gyrus tissue female adult (90 or above years))
+- Cell slim(s): None
+- Organ slim(s): brain
+- Developmental slim(s): ectoderm
+- System slim(s): central-nervous-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.ENCSR590EFT/logfile.modelling.fold_0.ENCSR590EFT.args.json

@@ -0,0 +1,45 @@
+{
+    "cmd": "pipeline",
+    "genome": "/scratch/groups/akundaje/anusri/chromatin_atlas/reference//hg38.genome.fa",
+    "chrom_sizes": "/scratch/groups/akundaje/anusri/chromatin_atlas/reference//chrom.sizes",
+    "input_bam_file": null,
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "bigwig": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE//ENCSR590EFT//preprocessing/bigWigs/ENCSR590EFT.bigWig",
+    "output_dir": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_0/chrombpnet_model/",
+    "data_type": "DNASE",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_0/chrombpnet_model/auxiliary/ENCSR590EFT_filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_0/chrombpnet_model/auxiliary/ENCSR590EFT_filtered.nonpeaks.bed",
+    "chr_fold_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/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/groups/akundaje/anusri/simg/chrombpnet_latest/chrombpnet/chrombpnet/training/models/chrombpnet_with_bias_model.py",
+    "file_prefix": "ENCSR590EFT",
+    "html_prefix": "./",
+    "bias_model_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_0/bias_model/models/ENCSR590EFT_bias.h5",
+    "negative_sampling_ratio": 0.1,
+    "filters": 512,
+    "n_dilation_layers": 8,
+    "max_jitter": 500,
+    "batch_size": 64,
+    "output_prefix": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_0/chrombpnet_model/models/ENCSR590EFT_chrombpnet",
+    "chr": "chr8",
+    "pwm_width": 24,
+    "params": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_0/chrombpnet_model/logs/ENCSR590EFT_chrombpnet_model_params.tsv"
+}

fold_0/logs.models.fold_0.ENCSR590EFT/logfile.modelling.fold_0.ENCSR590EFT.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR590EFT/failed_models_retrained/chrombpnet_model_feb15_fold_0/chrombpnet_model/models/ENCSR590EFT_bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR590EFT/fold_0/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR590EFT/logfile.modelling.fold_0.ENCSR590EFT.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	2850.18
+trainings_pts_post_thresh	173898

fold_0/logs.models.fold_0.ENCSR590EFT/logfile.modelling.fold_0.ENCSR590EFT.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR590EFT/failed_models_retrained/chrombpnet_model_feb15_fold_0/chrombpnet_model/models/ENCSR590EFT_chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR590EFT/fold_0/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR590EFT/logfile.modelling.fold_0.ENCSR590EFT.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	12.9
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_0/chrombpnet_model/models/ENCSR590EFT_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.ENCSR590EFT/logfile.modelling.fold_0.ENCSR590EFT.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR590EFT/failed_models_retrained/chrombpnet_model_feb15_fold_0/chrombpnet_model/models/ENCSR590EFT_chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR590EFT/fold_0/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR590EFT/logfile.modelling.fold_0.ENCSR590EFT.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
+0,0.43810197710990906,428.38427734375,434.03533935546875,0.17717397212982178,430.54290771484375,432.82843017578125
+1,0.19336125254631042,422.0543212890625,424.5483703613281,0.14884357154369354,429.5534973144531,431.4734191894531
+2,0.1754922866821289,420.44671630859375,422.7098083496094,0.14508463442325592,430.7346496582031,432.6061706542969
+3,0.16802430152893066,418.775634765625,420.9426574707031,0.1504591405391693,429.413818359375,431.3551025390625
+4,0.16110703349113464,417.8670349121094,419.9448547363281,0.14200018346309662,429.2088928222656,431.04052734375
+5,0.15702641010284424,417.3831787109375,419.4093322753906,0.13272283971309662,430.46112060546875,432.1737060546875
+6,0.1523531675338745,416.73443603515625,418.6993103027344,0.13206759095191956,429.27166748046875,430.9757080078125
+7,0.14877255260944366,416.2842712402344,418.2038879394531,0.1280241757631302,428.7884826660156,430.44024658203125
+8,0.1473131626844406,415.4502868652344,417.35064697265625,0.13264407217502594,430.8233642578125,432.5346374511719
+9,0.14397330582141876,415.0677795410156,416.9253234863281,0.1332000195980072,428.4639892578125,430.1821594238281
+10,0.14213836193084717,414.22137451171875,416.0548400878906,0.1271737813949585,430.9975891113281,432.63812255859375
+11,0.1405624896287918,414.102294921875,415.9154052734375,0.13292688131332397,430.4233703613281,432.1379089355469
+12,0.13854515552520752,413.5889587402344,415.3767395019531,0.12685392796993256,431.3756103515625,433.01190185546875
+13,0.13801176846027374,412.67254638671875,414.4532470703125,0.12944987416267395,430.5341491699219,432.2039794921875
+14,0.13674604892730713,412.659423828125,414.4240417480469,0.14037984609603882,432.2200622558594,434.0309143066406

fold_0/model.bias_scaled.fold_0.ENCSR590EFT.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f40c4b0b9da070787e8c456ca2563fa6f9e790d8be84e9cb0dfba6923088baa5
+size 2691928

fold_0/model.bias_scaled.fold_0.ENCSR590EFT.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9bdf70e1f6a3ca33acc0d016aaa66df5e42b71c8e6bd6a0a26e3fbc99d1767a
+size 1198080

fold_0/model.chrombpnet.fold_0.ENCSR590EFT.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47c1a1f9963d1332b6acd61fa64d36052b6a7c7deae278f126869a25bcd7b60b
+size 77538904

fold_0/model.chrombpnet.fold_0.ENCSR590EFT.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b078747733e6a93c1aaeae957849c0a5182ca5d73480b7b2a29ec46c7cca625d
+size 27525120

fold_0/model.chrombpnet_nobias.fold_0.ENCSR590EFT.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c973d792f2c8bd594cb5c7d02bece00bdd18bfd58ab2a40d694d1a7412c5dd3e
+size 25582648

fold_0/model.chrombpnet_nobias.fold_0.ENCSR590EFT.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:597c40e82e9a30347e12d1aec785df9dacd28a8921b6176bf85a11c1a3e0e973
+size 26060800

fold_1/logs.models.fold_1.ENCSR590EFT/logfile.modelling.fold_1.ENCSR590EFT.args.json

@@ -0,0 +1,45 @@
+{
+    "cmd": "train",
+    "genome": "/scratch/groups/akundaje/anusri/chromatin_atlas/reference//hg38.genome.fa",
+    "chrom_sizes": "/scratch/groups/akundaje/anusri/chromatin_atlas/reference//chrom.sizes",
+    "input_bam_file": null,
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "bigwig": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE//ENCSR590EFT//preprocessing/bigWigs/ENCSR590EFT.bigWig",
+    "output_dir": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_1/chrombpnet_model/",
+    "data_type": "DNASE",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_1/chrombpnet_model/auxiliary/ENCSR590EFT_filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_1/chrombpnet_model/auxiliary/ENCSR590EFT_filtered.nonpeaks.bed",
+    "chr_fold_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/splits/fold_1.json",
+    "outlier_threshold": 0.9999,
+    "ATAC_ref_path": null,
+    "DNASE_ref_path": null,
+    "num_samples": 10000,
+    "inputlen": 2114,
+    "outputlen": 1000,
+    "seed": 1234,
+    "epochs": 50,
+    "early_stop": 5,
+    "learning_rate": 0.001,
+    "trackables": [
+        "logcount_predictions_loss",
+        "loss",
+        "logits_profile_predictions_loss",
+        "val_logcount_predictions_loss",
+        "val_loss",
+        "val_logits_profile_predictions_loss"
+    ],
+    "architecture_from_file": "/home/groups/akundaje/anusri/simg/chrombpnet_latest/chrombpnet/chrombpnet/training/models/chrombpnet_with_bias_model.py",
+    "file_prefix": "ENCSR590EFT",
+    "html_prefix": "./",
+    "bias_model_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_0/bias_model/models/ENCSR590EFT_bias.h5",
+    "negative_sampling_ratio": 0.1,
+    "filters": 512,
+    "n_dilation_layers": 8,
+    "max_jitter": 500,
+    "batch_size": 64,
+    "output_prefix": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_1/chrombpnet_model/models/ENCSR590EFT_chrombpnet",
+    "chr": "chr12",
+    "pwm_width": 24,
+    "params": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_1/chrombpnet_model/logs/ENCSR590EFT_chrombpnet_model_params.tsv"
+}

fold_1/logs.models.fold_1.ENCSR590EFT/logfile.modelling.fold_1.ENCSR590EFT.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR590EFT/failed_models_retrained/chrombpnet_model_feb15_fold_1/chrombpnet_model/models/ENCSR590EFT_bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR590EFT/fold_1/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR590EFT/logfile.modelling.fold_1.ENCSR590EFT.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	2787.65
+trainings_pts_post_thresh	175225

fold_1/logs.models.fold_1.ENCSR590EFT/logfile.modelling.fold_1.ENCSR590EFT.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR590EFT/failed_models_retrained/chrombpnet_model_feb15_fold_1/chrombpnet_model/models/ENCSR590EFT_chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR590EFT/fold_1/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR590EFT/logfile.modelling.fold_1.ENCSR590EFT.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	12.9
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_1/chrombpnet_model/models/ENCSR590EFT_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.ENCSR590EFT/logfile.modelling.fold_1.ENCSR590EFT.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR590EFT/failed_models_retrained/chrombpnet_model_feb15_fold_1/chrombpnet_model/models/ENCSR590EFT_chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR590EFT/fold_1/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR590EFT/logfile.modelling.fold_1.ENCSR590EFT.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
+0,0.6603243947029114,425.78936767578125,434.3078308105469,0.20189890265464783,480.0959167480469,482.7002868652344
+1,0.2027665227651596,417.6494445800781,420.26519775390625,0.20290623605251312,476.2923278808594,478.9095153808594
+2,0.1859298199415207,415.7030029296875,418.10101318359375,0.16948571801185608,475.7789306640625,477.96539306640625
+3,0.17667917907238007,414.3038330078125,416.5835266113281,0.15979403257369995,472.4058532714844,474.4669494628906
+4,0.16704705357551575,413.404541015625,415.5593566894531,0.1462777704000473,472.46282958984375,474.3499450683594
+5,0.16016340255737305,413.0187072753906,415.0849609375,0.14491991698741913,470.9291076660156,472.7986145019531
+6,0.15577332675457,411.3818359375,413.3922424316406,0.15136633813381195,472.3423156738281,474.2947998046875
+7,0.15014496445655823,410.705810546875,412.64208984375,0.1423153132200241,473.85699462890625,475.6924743652344
+8,0.14856284856796265,410.5835876464844,412.4997253417969,0.15024793148040771,473.48114013671875,475.4193420410156
+9,0.1462765783071518,410.1086730957031,411.9956970214844,0.1374790519475937,474.7830505371094,476.55645751953125
+10,0.14371779561042786,408.8878479003906,410.7419738769531,0.13677054643630981,472.7637939453125,474.5281677246094

fold_1/model.bias_scaled.fold_1.ENCSR590EFT.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:678265952568caed2fd7d2554ce9d0b956d39576c743c22a5bb650afb5107743
+size 2691928

fold_1/model.bias_scaled.fold_1.ENCSR590EFT.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e75f0957eb1c8fd4f65e508dbfeef6c1964f6bac0396b0121eb39d414943bb99
+size 1198080

fold_1/model.chrombpnet.fold_1.ENCSR590EFT.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d9d414de870615c914d4040d341b6ab5d518a4ba772f78e71bee6e99eb0a27b
+size 77538904

fold_1/model.chrombpnet.fold_1.ENCSR590EFT.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:287ce2d7dab7563746f866bba6a0e805196d6a8a003dfd199b3d6f777d188d84
+size 27525120

fold_1/model.chrombpnet_nobias.fold_1.ENCSR590EFT.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9f01bdd2ab1232f3f8a673f22cf1481c1bd0f0b6265490a5b23e1a6b0c1f2cb2
+size 25582648

fold_1/model.chrombpnet_nobias.fold_1.ENCSR590EFT.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75a02fb87fe283413cb8d6a97ebc3b3e37d906f7452c497c6c4a4e7870222211
+size 26060800

fold_2/logs.models.fold_2.ENCSR590EFT/logfile.modelling.fold_2.ENCSR590EFT.args.json

@@ -0,0 +1,45 @@
+{
+    "cmd": "train",
+    "genome": "/scratch/groups/akundaje/anusri/chromatin_atlas/reference//hg38.genome.fa",
+    "chrom_sizes": "/scratch/groups/akundaje/anusri/chromatin_atlas/reference//chrom.sizes",
+    "input_bam_file": null,
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "bigwig": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE//ENCSR590EFT//preprocessing/bigWigs/ENCSR590EFT.bigWig",
+    "output_dir": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_2/chrombpnet_model/",
+    "data_type": "DNASE",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_2/chrombpnet_model/auxiliary/ENCSR590EFT_filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_2/chrombpnet_model/auxiliary/ENCSR590EFT_filtered.nonpeaks.bed",
+    "chr_fold_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/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/groups/akundaje/anusri/simg/chrombpnet_latest/chrombpnet/chrombpnet/training/models/chrombpnet_with_bias_model.py",
+    "file_prefix": "ENCSR590EFT",
+    "html_prefix": "./",
+    "bias_model_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_0/bias_model/models/ENCSR590EFT_bias.h5",
+    "negative_sampling_ratio": 0.1,
+    "filters": 512,
+    "n_dilation_layers": 8,
+    "max_jitter": 500,
+    "batch_size": 64,
+    "output_prefix": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_2/chrombpnet_model/models/ENCSR590EFT_chrombpnet",
+    "chr": "chr22",
+    "pwm_width": 24,
+    "params": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_2/chrombpnet_model/logs/ENCSR590EFT_chrombpnet_model_params.tsv"
+}

fold_2/logs.models.fold_2.ENCSR590EFT/logfile.modelling.fold_2.ENCSR590EFT.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR590EFT/failed_models_retrained/chrombpnet_model_feb15_fold_2/chrombpnet_model/models/ENCSR590EFT_bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR590EFT/fold_2/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR590EFT/logfile.modelling.fold_2.ENCSR590EFT.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	2833.26
+trainings_pts_post_thresh	178632

fold_2/logs.models.fold_2.ENCSR590EFT/logfile.modelling.fold_2.ENCSR590EFT.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR590EFT/failed_models_retrained/chrombpnet_model_feb15_fold_2/chrombpnet_model/models/ENCSR590EFT_chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR590EFT/fold_2/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR590EFT/logfile.modelling.fold_2.ENCSR590EFT.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	12.9
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_2/chrombpnet_model/models/ENCSR590EFT_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.ENCSR590EFT/logfile.modelling.fold_2.ENCSR590EFT.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR590EFT/failed_models_retrained/chrombpnet_model_feb15_fold_2/chrombpnet_model/models/ENCSR590EFT_chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR590EFT/fold_2/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR590EFT/logfile.modelling.fold_2.ENCSR590EFT.epoch_loss.csv

@@ -0,0 +1,10 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,0.7111934423446655,429.3876647949219,438.5624084472656,0.2829739451408386,464.3855895996094,468.03607177734375
+1,0.1980172097682953,421.21112060546875,423.7652893066406,0.18554885685443878,459.5437927246094,461.9375915527344
+2,0.18049883842468262,419.3358459472656,421.66461181640625,0.17580938339233398,457.4783630371094,459.7462463378906
+3,0.16695210337638855,417.856201171875,420.010009765625,0.16742119193077087,456.99951171875,459.15911865234375
+4,0.16156646609306335,416.9227600097656,419.0077819824219,0.1629873365163803,457.3108825683594,459.4134826660156
+5,0.1543513536453247,415.34649658203125,417.3373107910156,0.166383296251297,457.1653137207031,459.31170654296875
+6,0.15026794373989105,415.1611633300781,417.099853515625,0.15756934881210327,457.6061706542969,459.6385803222656
+7,0.1443234235048294,414.84527587890625,416.70721435546875,0.1641112118959427,458.906982421875,461.02435302734375
+8,0.14214248955249786,414.074462890625,415.9082336425781,0.17782342433929443,458.93487548828125,461.22882080078125

fold_2/model.bias_scaled.fold_2.ENCSR590EFT.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25a0719ed82d55b21a44a03f9dccb3f42012a31c07977cd9dde9dd3a7c9fc701
+size 2691928

fold_2/model.bias_scaled.fold_2.ENCSR590EFT.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87a5ffbf8ff0af532c9713372c88bb2c41761dd651a6fcdbebbebb7c62683c24
+size 1198080

fold_2/model.chrombpnet.fold_2.ENCSR590EFT.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:830e1e84dc4a74e8e81afd83a589227d92e4a769a1c116dcfa094e0a5804b43d
+size 77538904

fold_2/model.chrombpnet.fold_2.ENCSR590EFT.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e19098e1e0c60e86b12e0f27102953678edad3447c414c06f64917e9e1354684
+size 27525120

fold_2/model.chrombpnet_nobias.fold_2.ENCSR590EFT.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5684087d2c754f7490620dbdb1c7cd0389e0179d817cb154aa9aa75efbff36d2
+size 25582648

fold_2/model.chrombpnet_nobias.fold_2.ENCSR590EFT.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a01c6d37a6496b919d14f897d84f7fb068e9c0d3ed061310bea40edc9a5e45c6
+size 26060800

fold_3/logs.models.fold_3.ENCSR590EFT/logfile.modelling.fold_3.ENCSR590EFT.args.json

@@ -0,0 +1,45 @@
+{
+    "cmd": "train",
+    "genome": "/scratch/groups/akundaje/anusri/chromatin_atlas/reference//hg38.genome.fa",
+    "chrom_sizes": "/scratch/groups/akundaje/anusri/chromatin_atlas/reference//chrom.sizes",
+    "input_bam_file": null,
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "bigwig": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE//ENCSR590EFT//preprocessing/bigWigs/ENCSR590EFT.bigWig",
+    "output_dir": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_3/chrombpnet_model/",
+    "data_type": "DNASE",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_3/chrombpnet_model/auxiliary/ENCSR590EFT_filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_3/chrombpnet_model/auxiliary/ENCSR590EFT_filtered.nonpeaks.bed",
+    "chr_fold_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/splits/fold_3.json",
+    "outlier_threshold": 0.9999,
+    "ATAC_ref_path": null,
+    "DNASE_ref_path": null,
+    "num_samples": 10000,
+    "inputlen": 2114,
+    "outputlen": 1000,
+    "seed": 1234,
+    "epochs": 50,
+    "early_stop": 5,
+    "learning_rate": 0.001,
+    "trackables": [
+        "logcount_predictions_loss",
+        "loss",
+        "logits_profile_predictions_loss",
+        "val_logcount_predictions_loss",
+        "val_loss",
+        "val_logits_profile_predictions_loss"
+    ],
+    "architecture_from_file": "/home/groups/akundaje/anusri/simg/chrombpnet_latest/chrombpnet/chrombpnet/training/models/chrombpnet_with_bias_model.py",
+    "file_prefix": "ENCSR590EFT",
+    "html_prefix": "./",
+    "bias_model_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_0/bias_model/models/ENCSR590EFT_bias.h5",
+    "negative_sampling_ratio": 0.1,
+    "filters": 512,
+    "n_dilation_layers": 8,
+    "max_jitter": 500,
+    "batch_size": 64,
+    "output_prefix": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_3/chrombpnet_model/models/ENCSR590EFT_chrombpnet",
+    "chr": "chr6",
+    "pwm_width": 24,
+    "params": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_3/chrombpnet_model/logs/ENCSR590EFT_chrombpnet_model_params.tsv"
+}

fold_3/logs.models.fold_3.ENCSR590EFT/logfile.modelling.fold_3.ENCSR590EFT.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR590EFT/failed_models_retrained/chrombpnet_model_feb15_fold_3/chrombpnet_model/models/ENCSR590EFT_bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR590EFT/fold_3/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_3/logs.models.fold_3.ENCSR590EFT/logfile.modelling.fold_3.ENCSR590EFT.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	2846.43
+trainings_pts_post_thresh	172487

fold_3/logs.models.fold_3.ENCSR590EFT/logfile.modelling.fold_3.ENCSR590EFT.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR590EFT/failed_models_retrained/chrombpnet_model_feb15_fold_3/chrombpnet_model/models/ENCSR590EFT_chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR590EFT/fold_3/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_3/logs.models.fold_3.ENCSR590EFT/logfile.modelling.fold_3.ENCSR590EFT.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	12.9
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR590EFT//chrombpnet_model_feb15_fold_3/chrombpnet_model/models/ENCSR590EFT_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.ENCSR590EFT/logfile.modelling.fold_3.ENCSR590EFT.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR590EFT/failed_models_retrained/chrombpnet_model_feb15_fold_3/chrombpnet_model/models/ENCSR590EFT_chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR590EFT/fold_3/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py