Upload folder using huggingface_hub

README.md

@@ -0,0 +1,120 @@
+---
+license: mit
+library_name: chrombpnet
+tags:
+  - encode
+  - chrombpnet
+  - chromatin-accessibility
+  - DNASE
+  - 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: DNASE in lung (ENCSR504KZE)
+- Model: ChromBPNet
+- Assay: DNASE-seq
+- Experiment: [ENCSR504KZE](https://www.encodeproject.org/experiments/ENCSR504KZE/)
+- Model annotation: [ENCSR430JNN](https://www.encodeproject.org/annotations/ENCSR430JNN/)
+- Biosample: lung (Full name: Homo sapiens lung tissue embryo (67 days))
+- 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.ENCSR504KZE/logfile.modelling.fold_0.ENCSR504KZE.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//ENCSR504KZE//preprocessing/bigWigs/ENCSR504KZE.bigWig",
+    "output_dir": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_0/chrombpnet_model/",
+    "data_type": "DNASE",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_0/chrombpnet_model/auxiliary/ENCSR504KZE_filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_0/chrombpnet_model/auxiliary/ENCSR504KZE_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": "ENCSR504KZE",
+    "html_prefix": "./",
+    "bias_model_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_0/bias_model/models/ENCSR504KZE_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/ENCSR504KZE//chrombpnet_model_feb20_fold_0/chrombpnet_model/models/ENCSR504KZE_chrombpnet",
+    "chr": "chr8",
+    "pwm_width": 24,
+    "params": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_0/chrombpnet_model/logs/ENCSR504KZE_chrombpnet_model_params.tsv"
+}

fold_0/logs.models.fold_0.ENCSR504KZE/logfile.modelling.fold_0.ENCSR504KZE.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR504KZE/failed_models_retrained/chrombpnet_model_feb20_fold_0/chrombpnet_model/models/ENCSR504KZE_bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR504KZE/fold_0/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR504KZE/logfile.modelling.fold_0.ENCSR504KZE.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	965.3
+trainings_pts_post_thresh	169582

fold_0/logs.models.fold_0.ENCSR504KZE/logfile.modelling.fold_0.ENCSR504KZE.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR504KZE/failed_models_retrained/chrombpnet_model_feb20_fold_0/chrombpnet_model/models/ENCSR504KZE_chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR504KZE/fold_0/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR504KZE/logfile.modelling.fold_0.ENCSR504KZE.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	2.9
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_0/chrombpnet_model/models/ENCSR504KZE_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.ENCSR504KZE/logfile.modelling.fold_0.ENCSR504KZE.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR504KZE/failed_models_retrained/chrombpnet_model_feb20_fold_0/chrombpnet_model/models/ENCSR504KZE_chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR504KZE/fold_0/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR504KZE/logfile.modelling.fold_0.ENCSR504KZE.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
+0,1.5141040086746216,158.1604461669922,162.55136108398438,0.673427164554596,155.22317504882812,157.1760711669922
+1,0.6841257214546204,152.28114318847656,154.26492309570312,0.6764643788337708,152.5111083984375,154.472900390625
+2,0.6249499320983887,150.69288635253906,152.50485229492188,0.5706433653831482,151.8504638671875,153.50538635253906
+3,0.5907095074653625,149.52035522460938,151.2336883544922,0.5806581377983093,152.29849243164062,153.982421875
+4,0.562157154083252,148.72903442382812,150.3590087890625,0.5742468237876892,151.2978057861328,152.96315002441406
+5,0.5395950078964233,148.08944702148438,149.6542205810547,0.5495128631591797,151.06838989257812,152.66197204589844
+6,0.5243784189224243,147.54904174804688,149.06971740722656,0.5259464383125305,151.37953186035156,152.9047088623047
+7,0.5113770365715027,146.89501953125,148.37791442871094,0.5827067494392395,150.8873748779297,152.57720947265625
+8,0.49797797203063965,146.28839111328125,147.73243713378906,0.5105103850364685,151.92813110351562,153.40869140625
+9,0.4925503432750702,145.9954071044922,147.42376708984375,0.5071589946746826,152.24539184570312,153.71604919433594
+10,0.4793643653392792,145.55787658691406,146.94776916503906,0.5024353265762329,152.04176330566406,153.49879455566406
+11,0.4754734933376312,145.02569580078125,146.40440368652344,0.5090529322624207,152.40936279296875,153.88563537597656
+12,0.46678003668785095,144.74810791015625,146.10182189941406,0.5248997807502747,152.10433959960938,153.62646484375

fold_0/model.bias_scaled.fold_0.ENCSR504KZE.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d0238c7c13e86ad93ba6fd4390ed574b398c11491a87b4585b563dfec084414
+size 2691928

fold_0/model.bias_scaled.fold_0.ENCSR504KZE.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8e37ca00cbfe9c7b01e405d7ea30376262212ad55dd162b0a78e1b9ad76ad1b
+size 1198080

fold_0/model.chrombpnet.fold_0.ENCSR504KZE.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e2405ea8c792a947253ea785e1a5f41ade89d48e8f7749bea69994e43982870
+size 77538904

fold_0/model.chrombpnet.fold_0.ENCSR504KZE.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00149ce434e0be7a5a6f3e707200a1edf99745726ffa1d85d60cb2a79f50df3a
+size 27525120

fold_0/model.chrombpnet_nobias.fold_0.ENCSR504KZE.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d46bd3305a3294cfa0f0c0f567876917431b8b1d68fe13f46414e3c780699bc4
+size 25582648

fold_0/model.chrombpnet_nobias.fold_0.ENCSR504KZE.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:639fbcda385585f214c7a1feaf03e636d6739d279ee697635e2f50638ac71ee1
+size 26060800

fold_1/logs.models.fold_1.ENCSR504KZE/logfile.modelling.fold_1.ENCSR504KZE.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//ENCSR504KZE//preprocessing/bigWigs/ENCSR504KZE.bigWig",
+    "output_dir": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_1/chrombpnet_model/",
+    "data_type": "DNASE",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_1/chrombpnet_model/auxiliary/ENCSR504KZE_filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_1/chrombpnet_model/auxiliary/ENCSR504KZE_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": "ENCSR504KZE",
+    "html_prefix": "./",
+    "bias_model_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_0/bias_model/models/ENCSR504KZE_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/ENCSR504KZE//chrombpnet_model_feb20_fold_1/chrombpnet_model/models/ENCSR504KZE_chrombpnet",
+    "chr": "chr12",
+    "pwm_width": 24,
+    "params": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_1/chrombpnet_model/logs/ENCSR504KZE_chrombpnet_model_params.tsv"
+}

fold_1/logs.models.fold_1.ENCSR504KZE/logfile.modelling.fold_1.ENCSR504KZE.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR504KZE/failed_models_retrained/chrombpnet_model_feb20_fold_1/chrombpnet_model/models/ENCSR504KZE_bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR504KZE/fold_1/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR504KZE/logfile.modelling.fold_1.ENCSR504KZE.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	969.52
+trainings_pts_post_thresh	172544

fold_1/logs.models.fold_1.ENCSR504KZE/logfile.modelling.fold_1.ENCSR504KZE.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR504KZE/failed_models_retrained/chrombpnet_model_feb20_fold_1/chrombpnet_model/models/ENCSR504KZE_chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR504KZE/fold_1/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR504KZE/logfile.modelling.fold_1.ENCSR504KZE.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	3.0
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_1/chrombpnet_model/models/ENCSR504KZE_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.ENCSR504KZE/logfile.modelling.fold_1.ENCSR504KZE.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR504KZE/failed_models_retrained/chrombpnet_model_feb20_fold_1/chrombpnet_model/models/ENCSR504KZE_chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR504KZE/fold_1/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR504KZE/logfile.modelling.fold_1.ENCSR504KZE.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
+0,3.522555112838745,159.528076171875,170.09573364257812,0.7973940968513489,179.5071258544922,181.8993682861328
+1,0.7348886728286743,152.9197540283203,155.1243896484375,0.7162854075431824,176.99464416503906,179.1433868408203
+2,0.6736692786216736,151.2919158935547,153.3130645751953,0.6294271945953369,175.65394592285156,177.54226684570312
+3,0.6277006268501282,149.75656127929688,151.63958740234375,0.6308123469352722,174.4501495361328,176.3428497314453
+4,0.5924352407455444,149.165283203125,150.94284057617188,0.6524657011032104,173.8167724609375,175.77410888671875
+5,0.5703325271606445,148.47251892089844,150.1833953857422,0.5499668121337891,173.34036254882812,174.990234375
+6,0.5552316904067993,147.73800659179688,149.40371704101562,0.5297204852104187,172.82029724121094,174.40948486328125
+7,0.5333635807037354,147.3335418701172,148.9337921142578,0.6952446699142456,174.24057006835938,176.3262176513672
+8,0.5213336944580078,146.76625061035156,148.33038330078125,0.5195837616920471,172.92984008789062,174.4886016845703
+9,0.5095585584640503,146.09854125976562,147.62698364257812,0.5131277441978455,173.11416625976562,174.65354919433594
+10,0.4951857626438141,145.81289672851562,147.29830932617188,0.5048381090164185,174.04840087890625,175.5628662109375
+11,0.49113476276397705,145.25576782226562,146.72906494140625,0.507943332195282,174.8149871826172,176.3386993408203

fold_1/model.bias_scaled.fold_1.ENCSR504KZE.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a9792d1baf24f9ee09adfa9eb690cf7d7835fe5d79a9ff645a978eae184bf7a
+size 2691928

fold_1/model.bias_scaled.fold_1.ENCSR504KZE.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d44f518e0c9fa403cc5404775feb49cd1ff0be7a579a9dff86c3a404504c1ea
+size 1198080

fold_1/model.chrombpnet.fold_1.ENCSR504KZE.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aaacb37ec95a9966773688b71d6010f2aa49b9721291f0893c99281acea5d4b3
+size 77538904

fold_1/model.chrombpnet.fold_1.ENCSR504KZE.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a02e1c9bbc8d0ea74acbcc5fb6566abe6bbd3fbd0a5153607cabee525047dfef
+size 27525120

fold_1/model.chrombpnet_nobias.fold_1.ENCSR504KZE.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2eb231c24b24c6fdca5693552d4195c49c6c8a4a93731194a479f910d1508838
+size 25582648

fold_1/model.chrombpnet_nobias.fold_1.ENCSR504KZE.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22520ef00a8cc5bb4b5ae667ba7818fc78fe1923207a5335137bf18bae80377a
+size 26060800

fold_2/logs.models.fold_2.ENCSR504KZE/logfile.modelling.fold_2.ENCSR504KZE.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//ENCSR504KZE//preprocessing/bigWigs/ENCSR504KZE.bigWig",
+    "output_dir": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_2/chrombpnet_model/",
+    "data_type": "DNASE",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_2/chrombpnet_model/auxiliary/ENCSR504KZE_filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_2/chrombpnet_model/auxiliary/ENCSR504KZE_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": "ENCSR504KZE",
+    "html_prefix": "./",
+    "bias_model_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_0/bias_model/models/ENCSR504KZE_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/ENCSR504KZE//chrombpnet_model_feb20_fold_2/chrombpnet_model/models/ENCSR504KZE_chrombpnet",
+    "chr": "chr22",
+    "pwm_width": 24,
+    "params": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_2/chrombpnet_model/logs/ENCSR504KZE_chrombpnet_model_params.tsv"
+}

fold_2/logs.models.fold_2.ENCSR504KZE/logfile.modelling.fold_2.ENCSR504KZE.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR504KZE/failed_models_retrained/chrombpnet_model_feb20_fold_2/chrombpnet_model/models/ENCSR504KZE_bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR504KZE/fold_2/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR504KZE/logfile.modelling.fold_2.ENCSR504KZE.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	958.34
+trainings_pts_post_thresh	177381

fold_2/logs.models.fold_2.ENCSR504KZE/logfile.modelling.fold_2.ENCSR504KZE.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR504KZE/failed_models_retrained/chrombpnet_model_feb20_fold_2/chrombpnet_model/models/ENCSR504KZE_chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR504KZE/fold_2/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR504KZE/logfile.modelling.fold_2.ENCSR504KZE.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	2.9
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_2/chrombpnet_model/models/ENCSR504KZE_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.ENCSR504KZE/logfile.modelling.fold_2.ENCSR504KZE.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR504KZE/failed_models_retrained/chrombpnet_model_feb20_fold_2/chrombpnet_model/models/ENCSR504KZE_chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR504KZE/fold_2/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR504KZE/logfile.modelling.fold_2.ENCSR504KZE.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
+0,2.643158197402954,159.48544311523438,167.15081787109375,0.7547848224639893,160.38241577148438,162.57130432128906
+1,0.7178019285202026,153.5948486328125,155.67640686035156,0.6639777421951294,158.0540771484375,159.97975158691406
+2,0.650472104549408,151.6879425048828,153.57437133789062,0.6315906643867493,156.9258575439453,158.75755310058594
+3,0.6129633188247681,150.43902587890625,152.2164764404297,0.5856024026870728,156.40431213378906,158.1026611328125
+4,0.5768957138061523,149.81427001953125,151.4870147705078,0.56443190574646,156.41079711914062,158.0475616455078
+5,0.5554076433181763,149.026611328125,150.63705444335938,0.5463960766792297,156.13795471191406,157.72250366210938
+6,0.535622239112854,148.64402770996094,150.19728088378906,0.5428170561790466,155.86834716796875,157.44244384765625
+7,0.5255836248397827,147.81884765625,149.34327697753906,0.5307384729385376,154.917236328125,156.45639038085938
+8,0.5094293355941772,147.4803924560547,148.9579620361328,0.5612806677818298,155.47665405273438,157.1042938232422
+9,0.4999590516090393,146.7877655029297,148.23745727539062,0.5282790064811707,155.96807861328125,157.50018310546875
+10,0.4900468587875366,146.47927856445312,147.90040588378906,0.528927206993103,156.0289764404297,157.562744140625
+11,0.4769517779350281,146.23825073242188,147.62149047851562,0.5514208674430847,156.76657104492188,158.36561584472656
+12,0.47620975971221924,145.6007080078125,146.98150634765625,0.510372519493103,156.9517059326172,158.4318084716797

fold_2/model.bias_scaled.fold_2.ENCSR504KZE.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c16d6a3dfe824e9da98618459bc0b46e4cf590ac91d06e04858ff0b28c0c90aa
+size 2691928

fold_2/model.bias_scaled.fold_2.ENCSR504KZE.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13e7abb8c7dbce03dcd1f29560d69b71a5eaffe79d29ce4d6aab07f7ed5d62ea
+size 1198080

fold_2/model.chrombpnet.fold_2.ENCSR504KZE.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e9586d2d3f1390a0db43fbdb88dd554480c83cf4b30cf616f455026e3586b20
+size 77538904

fold_2/model.chrombpnet.fold_2.ENCSR504KZE.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ac90cc1f104a5952389163c19f79587104040d4180b32473e23a92fb28adef4
+size 27525120

fold_2/model.chrombpnet_nobias.fold_2.ENCSR504KZE.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8be1429e5a6d7b68ce8e42a843ac3acf3b44aa02dfe4244c7b21eba0511a679
+size 25582648

fold_2/model.chrombpnet_nobias.fold_2.ENCSR504KZE.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:125a23e138243913fb2627eb76219cf737c64149f2167f7db30fff96b47e4495
+size 26060800

fold_3/logs.models.fold_3.ENCSR504KZE/logfile.modelling.fold_3.ENCSR504KZE.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//ENCSR504KZE//preprocessing/bigWigs/ENCSR504KZE.bigWig",
+    "output_dir": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_3/chrombpnet_model/",
+    "data_type": "DNASE",
+    "peaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_3/chrombpnet_model/auxiliary/ENCSR504KZE_filtered.peaks.bed",
+    "nonpeaks": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_3/chrombpnet_model/auxiliary/ENCSR504KZE_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": "ENCSR504KZE",
+    "html_prefix": "./",
+    "bias_model_path": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_0/bias_model/models/ENCSR504KZE_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/ENCSR504KZE//chrombpnet_model_feb20_fold_3/chrombpnet_model/models/ENCSR504KZE_chrombpnet",
+    "chr": "chr6",
+    "pwm_width": 24,
+    "params": "/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_3/chrombpnet_model/logs/ENCSR504KZE_chrombpnet_model_params.tsv"
+}

fold_3/logs.models.fold_3.ENCSR504KZE/logfile.modelling.fold_3.ENCSR504KZE.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR504KZE/failed_models_retrained/chrombpnet_model_feb20_fold_3/chrombpnet_model/models/ENCSR504KZE_bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR504KZE/fold_3/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_3/logs.models.fold_3.ENCSR504KZE/logfile.modelling.fold_3.ENCSR504KZE.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	967.61
+trainings_pts_post_thresh	171569

fold_3/logs.models.fold_3.ENCSR504KZE/logfile.modelling.fold_3.ENCSR504KZE.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR504KZE/failed_models_retrained/chrombpnet_model_feb20_fold_3/chrombpnet_model/models/ENCSR504KZE_chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR504KZE/fold_3/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_3/logs.models.fold_3.ENCSR504KZE/logfile.modelling.fold_3.ENCSR504KZE.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	3.0
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/groups/akundaje/anusri/chromatin_atlas/DNASE/ENCSR504KZE//chrombpnet_model_feb20_fold_3/chrombpnet_model/models/ENCSR504KZE_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.ENCSR504KZE/logfile.modelling.fold_3.ENCSR504KZE.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/DNASE/ENCSR504KZE/failed_models_retrained/chrombpnet_model_feb20_fold_3/chrombpnet_model/models/ENCSR504KZE_chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/DNASE/ENCSR504KZE/fold_3/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py