Upload folder using huggingface_hub

README.md

@@ -0,0 +1,120 @@
+---
+license: mit
+library_name: chrombpnet
+tags:
+  - encode
+  - chrombpnet
+  - chromatin-accessibility
+  - ATAC
+  - heart
+  - 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 heart right ventricle (ENCSR403GBR)
+- Model: ChromBPNet
+- Assay: ATAC-seq
+- Experiment: [ENCSR403GBR](https://www.encodeproject.org/experiments/ENCSR403GBR/)
+- Model annotation: [ENCSR311LQE](https://www.encodeproject.org/annotations/ENCSR311LQE/)
+- Biosample: heart right ventricle (Full name: Homo sapiens heart right ventricle tissue male adult (55 years))
+- Cell slim(s): None
+- Organ slim(s): heart
+- Developmental slim(s): mesoderm
+- System slim(s): circulatory-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.ENCSR403GBR/logfile.modelling.fold_0.ENCSR403GBR.args.json

@@ -0,0 +1,50 @@
+{
+    "cmd": "pipeline",
+    "genome": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_no_alt_analysis_set_GCA_000001405.15.fasta",
+    "chrom_sizes": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_EBV.chrom.sizes.tsv",
+    "input_bam_file": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/preprocessing/bigWigs/ENCSR403GBR.bigWig",
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "output_dir": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/model_outputs",
+    "data_type": "ATAC",
+    "peaks": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/model_outputs/auxiliary/filtered.peaks.bed",
+    "nonpeaks": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/model_outputs/auxiliary/filtered.nonpeaks.bed",
+    "chr_fold_path": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/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/vhecht/chrombpnet/chrombpnet/chrombpnet/training/models/chrombpnet_with_bias_model.py",
+    "file_prefix": null,
+    "html_prefix": "./",
+    "bsort": false,
+    "tmpdir": null,
+    "no_st": false,
+    "bias_model_path": "/scratch/users/vhecht/chromatin-atlas-2022/reference/ATAC/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": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/model_outputs/models/chrombpnet",
+    "bigwig": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/model_outputs/auxiliary/data_unstranded.bw",
+    "plus_shift": null,
+    "minus_shift": null,
+    "chr": "chr8",
+    "pwm_width": 24,
+    "params": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/model_outputs/logs/chrombpnet_model_params.tsv"
+}

fold_0/logs.models.fold_0.ENCSR403GBR/logfile.modelling.fold_0.ENCSR403GBR.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/model_outputs/models/bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR403GBR/logfile.modelling.fold_0.ENCSR403GBR.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	5611.99
+trainings_pts_post_thresh	148719

fold_0/logs.models.fold_0.ENCSR403GBR/logfile.modelling.fold_0.ENCSR403GBR.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/model_outputs/models/chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR403GBR/logfile.modelling.fold_0.ENCSR403GBR.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	32.7
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/model_outputs/models/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_0.json
+negative_sampling_ratio	0.1

fold_0/logs.models.fold_0.ENCSR403GBR/logfile.modelling.fold_0.ENCSR403GBR.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/model_outputs/models/chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_0/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_0/logs.models.fold_0.ENCSR403GBR/logfile.modelling.fold_0.ENCSR403GBR.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
+0,1.1747137308120728,545.7109985351562,584.1239013671875,0.2846543490886688,539.9358520507812,549.2442016601562
+1,0.2508697509765625,532.6647338867188,540.8684692382812,0.21483151614665985,531.4309692382812,538.4558715820312
+2,0.2241925150156021,527.8343505859375,535.166015625,0.1826147735118866,529.3046264648438,535.2763671875
+3,0.21043893694877625,524.8626098632812,531.74462890625,0.172647163271904,528.800537109375,534.4461059570312
+4,0.1986246109008789,522.2210693359375,528.7164306640625,0.16763530671596527,527.5762939453125,533.0579223632812
+5,0.18767127394676208,519.74169921875,525.8783569335938,0.1692759096622467,527.1959228515625,532.7311401367188
+6,0.17515906691551208,517.53515625,523.2628784179688,0.16429118812084198,527.728515625,533.1007080078125
+7,0.1737983524799347,515.99951171875,521.68212890625,0.15648123621940613,527.5568237304688,532.6734008789062
+8,0.16482599079608917,514.0278930664062,519.4174194335938,0.1554720401763916,528.1880493164062,533.2720336914062
+9,0.16040001809597015,513.2587890625,518.50390625,0.1542883813381195,527.6629638671875,532.7080688476562
+10,0.15360009670257568,512.0770874023438,517.10009765625,0.14902393519878387,528.03466796875,532.9077758789062
+11,0.14766795933246613,510.8023986816406,515.6311645507812,0.15578962862491608,528.079345703125,533.1734619140625
+12,0.14514321088790894,510.25030517578125,514.9966430664062,0.15325024724006653,528.8095703125,533.8207397460938

fold_0/model.bias_scaled.fold_0.ENCSR403GBR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de0a30aa4685adc8a84cd35611e7f838f3fea29e17dd10cd385ccb7d5ac203d6
+size 2691928

fold_0/model.bias_scaled.fold_0.ENCSR403GBR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:554f95ac7b098c29a97c05ab2911dea822ac2437ecdaae4e9e76594682bac0ca
+size 1198080

fold_0/model.chrombpnet.fold_0.ENCSR403GBR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3799a2aa1aeb94338226e272240c43a19f955a4df214541f2931747ac75c1acf
+size 77538896

fold_0/model.chrombpnet.fold_0.ENCSR403GBR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:988541844d16c3219b0b6199f3184df9899c5044f5dde8b041bd8695ce73a132
+size 27535360

fold_0/model.chrombpnet_nobias.fold_0.ENCSR403GBR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e78a05d8ced3e5743d0adac629edc00f974c23d570d297e2954c4352e1d6a9b
+size 25582648

fold_0/model.chrombpnet_nobias.fold_0.ENCSR403GBR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76f2728e9313e079e6086e8dc822bd0a76a35b2dbb5779e2b9886f15506082d3
+size 26060800

fold_1/logs.models.fold_1.ENCSR403GBR/logfile.modelling.fold_1.ENCSR403GBR.args.json

@@ -0,0 +1,50 @@
+{
+    "cmd": "pipeline",
+    "genome": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_no_alt_analysis_set_GCA_000001405.15.fasta",
+    "chrom_sizes": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_EBV.chrom.sizes.tsv",
+    "input_bam_file": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/preprocessing/bigWigs/ENCSR403GBR.bigWig",
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "output_dir": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/model_outputs",
+    "data_type": "ATAC",
+    "peaks": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/model_outputs/auxiliary/filtered.peaks.bed",
+    "nonpeaks": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/model_outputs/auxiliary/filtered.nonpeaks.bed",
+    "chr_fold_path": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/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/users/vhecht/chrombpnet/chrombpnet/chrombpnet/training/models/chrombpnet_with_bias_model.py",
+    "file_prefix": null,
+    "html_prefix": "./",
+    "bsort": false,
+    "tmpdir": null,
+    "no_st": false,
+    "bias_model_path": "/scratch/users/vhecht/chromatin-atlas-2022/reference/ATAC/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": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/model_outputs/models/chrombpnet",
+    "bigwig": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/model_outputs/auxiliary/data_unstranded.bw",
+    "plus_shift": null,
+    "minus_shift": null,
+    "chr": "chr12",
+    "pwm_width": 24,
+    "params": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/model_outputs/logs/chrombpnet_model_params.tsv"
+}

fold_1/logs.models.fold_1.ENCSR403GBR/logfile.modelling.fold_1.ENCSR403GBR.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/model_outputs/models/bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR403GBR/logfile.modelling.fold_1.ENCSR403GBR.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	5372.12
+trainings_pts_post_thresh	150710

fold_1/logs.models.fold_1.ENCSR403GBR/logfile.modelling.fold_1.ENCSR403GBR.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/model_outputs/models/chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR403GBR/logfile.modelling.fold_1.ENCSR403GBR.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	32.7
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/model_outputs/models/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_1.json
+negative_sampling_ratio	0.1

fold_1/logs.models.fold_1.ENCSR403GBR/logfile.modelling.fold_1.ENCSR403GBR.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/model_outputs/models/chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_1/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_1/logs.models.fold_1.ENCSR403GBR/logfile.modelling.fold_1.ENCSR403GBR.epoch_loss.csv

@@ -0,0 +1,17 @@
+epoch,logcount_predictions_loss,logits_profile_predictions_loss,loss,val_logcount_predictions_loss,val_logits_profile_predictions_loss,val_loss
+0,1.5691070556640625,543.7957763671875,595.1061401367188,0.3288375735282898,591.0797119140625,601.8330078125
+1,0.24712030589580536,532.2855834960938,540.3663330078125,0.2145545482635498,579.2717895507812,586.287841796875
+2,0.22399336099624634,525.79638671875,533.1211547851562,0.27944818139076233,576.1354370117188,585.2735595703125
+3,0.21231012046337128,522.6170654296875,529.56005859375,0.21425744891166687,574.0201416015625,581.0259399414062
+4,0.2059667408466339,519.7216796875,526.4562377929688,0.20790013670921326,572.3535766601562,579.1519775390625
+5,0.19387312233448029,517.6648559570312,524.0042114257812,0.19231286644935608,571.369873046875,577.6583862304688
+6,0.1837160885334015,515.6544799804688,521.66162109375,0.17848438024520874,571.1610717773438,576.9977416992188
+7,0.17968948185443878,513.1287841796875,519.0049438476562,0.1874190866947174,571.3007202148438,577.4293212890625
+8,0.16710227727890015,511.7466125488281,517.2111206054688,0.1678396463394165,571.0838623046875,576.5722045898438
+9,0.1631269007921219,509.9712219238281,515.3048706054688,0.24546252191066742,571.814453125,579.8412475585938
+10,0.15701137483119965,508.650390625,513.7842407226562,0.16528798639774323,571.0624389648438,576.4671630859375
+11,0.1529230922460556,507.14495849609375,512.1460571289062,0.17551399767398834,570.784423828125,576.5238037109375
+12,0.14784842729568481,506.6151123046875,511.4493408203125,0.1797078400850296,571.1715087890625,577.0480346679688
+13,0.1417769491672516,505.7057800292969,510.3418884277344,0.165986105799675,572.12890625,577.5565185546875
+14,0.13824567198753357,504.8712463378906,509.39166259765625,0.1623615026473999,571.5148315429688,576.8241577148438
+15,0.1336216926574707,504.5464172363281,508.91546630859375,0.1655711829662323,573.3250122070312,578.7384643554688

fold_1/model.bias_scaled.fold_1.ENCSR403GBR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a821f58650b0538d1f4db7296ea65abb03efe8b4997e5c1823625db0bdaa7d0f
+size 2691928

fold_1/model.bias_scaled.fold_1.ENCSR403GBR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20937492e0c011665584fe2779638f9f3ae1185fd450b140b80cd845ce0d9ee5
+size 1198080

fold_1/model.chrombpnet.fold_1.ENCSR403GBR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62a6bb8eace1c02a90c5989574f64d143ec8c3bb9b9561a2307d69683ab33750
+size 77538896

fold_1/model.chrombpnet.fold_1.ENCSR403GBR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:277d106ea1338a8baee6b3b5b4fa390ebda7fa3bb5b1e307ca8771c3490d6024
+size 27535360

fold_1/model.chrombpnet_nobias.fold_1.ENCSR403GBR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16acf192c67620467346bc1f7667b4a66d39b7e71658915aba3c8051496b7f48
+size 25582648

fold_1/model.chrombpnet_nobias.fold_1.ENCSR403GBR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60e77922eaff946b83cf9049f33237d943b71aaad38543e5dcd2139ab5d0e283
+size 26060800

fold_2/logs.models.fold_2.ENCSR403GBR/logfile.modelling.fold_2.ENCSR403GBR.args.json

@@ -0,0 +1,50 @@
+{
+    "cmd": "pipeline",
+    "genome": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_no_alt_analysis_set_GCA_000001405.15.fasta",
+    "chrom_sizes": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_EBV.chrom.sizes.tsv",
+    "input_bam_file": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/preprocessing/bigWigs/ENCSR403GBR.bigWig",
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "output_dir": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/model_outputs",
+    "data_type": "ATAC",
+    "peaks": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/model_outputs/auxiliary/filtered.peaks.bed",
+    "nonpeaks": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/model_outputs/auxiliary/filtered.nonpeaks.bed",
+    "chr_fold_path": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_2.json",
+    "outlier_threshold": 0.9999,
+    "ATAC_ref_path": null,
+    "DNASE_ref_path": null,
+    "num_samples": 10000,
+    "inputlen": 2114,
+    "outputlen": 1000,
+    "seed": 1234,
+    "epochs": 50,
+    "early_stop": 5,
+    "learning_rate": 0.001,
+    "trackables": [
+        "logcount_predictions_loss",
+        "loss",
+        "logits_profile_predictions_loss",
+        "val_logcount_predictions_loss",
+        "val_loss",
+        "val_logits_profile_predictions_loss"
+    ],
+    "architecture_from_file": "/home/users/vhecht/chrombpnet/chrombpnet/chrombpnet/training/models/chrombpnet_with_bias_model.py",
+    "file_prefix": null,
+    "html_prefix": "./",
+    "bsort": false,
+    "tmpdir": null,
+    "no_st": false,
+    "bias_model_path": "/scratch/users/vhecht/chromatin-atlas-2022/reference/ATAC/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": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/model_outputs/models/chrombpnet",
+    "bigwig": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/model_outputs/auxiliary/data_unstranded.bw",
+    "plus_shift": null,
+    "minus_shift": null,
+    "chr": "chr22",
+    "pwm_width": 24,
+    "params": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/model_outputs/logs/chrombpnet_model_params.tsv"
+}

fold_2/logs.models.fold_2.ENCSR403GBR/logfile.modelling.fold_2.ENCSR403GBR.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/model_outputs/models/bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR403GBR/logfile.modelling.fold_2.ENCSR403GBR.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	5433.73
+trainings_pts_post_thresh	155306

fold_2/logs.models.fold_2.ENCSR403GBR/logfile.modelling.fold_2.ENCSR403GBR.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/model_outputs/models/chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR403GBR/logfile.modelling.fold_2.ENCSR403GBR.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	32.6
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/model_outputs/models/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_2.json
+negative_sampling_ratio	0.1

fold_2/logs.models.fold_2.ENCSR403GBR/logfile.modelling.fold_2.ENCSR403GBR.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/model_outputs/models/chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_2/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py

fold_2/logs.models.fold_2.ENCSR403GBR/logfile.modelling.fold_2.ENCSR403GBR.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
+0,1.0065038204193115,545.2531127929688,578.0653686523438,0.28043895959854126,554.2754516601562,563.4174194335938
+1,0.23452764749526978,530.9881591796875,538.63427734375,0.21513453125953674,544.88232421875,551.8956909179688
+2,0.2157915234565735,526.4489135742188,533.4837646484375,0.2089405208826065,542.8514404296875,549.6630249023438
+3,0.19656476378440857,523.3472290039062,529.7550048828125,0.20024855434894562,542.0309448242188,548.55908203125
+4,0.18607094883918762,520.6064453125,526.6714477539062,0.1987304389476776,540.6126098632812,547.0911865234375
+5,0.17968258261680603,518.0722045898438,523.9306030273438,0.18787448108196259,540.3778076171875,546.5026245117188
+6,0.1691441833972931,516.292236328125,521.8065795898438,0.1830214411020279,540.5828857421875,546.5493774414062
+7,0.1627778261899948,514.3131103515625,519.6187744140625,0.20049181580543518,541.6527709960938,548.188720703125
+8,0.15666961669921875,513.0729370117188,518.1810302734375,0.18216994404792786,540.91845703125,546.8573608398438
+9,0.15095078945159912,511.7507019042969,516.6716918945312,0.1845192313194275,540.7803955078125,546.7957153320312
+10,0.1481981873512268,510.5426940917969,515.3740844726562,0.18066509068012238,540.7384643554688,546.6282958984375

fold_2/model.bias_scaled.fold_2.ENCSR403GBR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3988e69b9e91e885145e65f10a81f83ae93b196d18c4003bd89795c91530e38f
+size 2691928

fold_2/model.bias_scaled.fold_2.ENCSR403GBR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d38fd659883806ce871088773996a3fe14f7e6fc5f39ea23177be1c0148ed41c
+size 1198080

fold_2/model.chrombpnet.fold_2.ENCSR403GBR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e76f2902549d5910e51499467d886d2374fc8c177a8171418e16ee3f7b7326e
+size 77538896

fold_2/model.chrombpnet.fold_2.ENCSR403GBR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9f9676ba2ec6d073319da803a6eaf42cf6d21fa2fb0274d27e6766ba655ed6a3
+size 27535360

fold_2/model.chrombpnet_nobias.fold_2.ENCSR403GBR.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48c21af544fb9fb38b54b1b811b55ab686c1fcf3bd2ad29bbc560895e6e087e9
+size 25582648

fold_2/model.chrombpnet_nobias.fold_2.ENCSR403GBR.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d69fca7bdd579426a237dfdad7492814cc5a1b4ebc04f20e3cb4f68b916744f9
+size 26060800

fold_3/logs.models.fold_3.ENCSR403GBR/logfile.modelling.fold_3.ENCSR403GBR.args.json

@@ -0,0 +1,50 @@
+{
+    "cmd": "pipeline",
+    "genome": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_no_alt_analysis_set_GCA_000001405.15.fasta",
+    "chrom_sizes": "/oak/stanford/groups/akundaje/ziwei75/atac_seq_pipeline/hg38/GRCh38_EBV.chrom.sizes.tsv",
+    "input_bam_file": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/preprocessing/bigWigs/ENCSR403GBR.bigWig",
+    "input_fragment_file": null,
+    "input_tagalign_file": null,
+    "output_dir": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/model_outputs",
+    "data_type": "ATAC",
+    "peaks": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/model_outputs/auxiliary/filtered.peaks.bed",
+    "nonpeaks": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/model_outputs/auxiliary/filtered.nonpeaks.bed",
+    "chr_fold_path": "/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/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/users/vhecht/chrombpnet/chrombpnet/chrombpnet/training/models/chrombpnet_with_bias_model.py",
+    "file_prefix": null,
+    "html_prefix": "./",
+    "bsort": false,
+    "tmpdir": null,
+    "no_st": false,
+    "bias_model_path": "/scratch/users/vhecht/chromatin-atlas-2022/reference/ATAC/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": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/model_outputs/models/chrombpnet",
+    "bigwig": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/model_outputs/auxiliary/data_unstranded.bw",
+    "plus_shift": null,
+    "minus_shift": null,
+    "chr": "chr6",
+    "pwm_width": 24,
+    "params": "/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/model_outputs/logs/chrombpnet_model_params.tsv"
+}

fold_3/logs.models.fold_3.ENCSR403GBR/logfile.modelling.fold_3.ENCSR403GBR.bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/model_outputs/models/bias_model_scaled.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/new_model_format/bias_model_scaled.tar with get_new_tf_model_format.py

fold_3/logs.models.fold_3.ENCSR403GBR/logfile.modelling.fold_3.ENCSR403GBR.chrombpnet_data_params.tsv

@@ -0,0 +1,3 @@
+counts_sum_min_thresh	0.0
+counts_sum_max_thresh	5601.92
+trainings_pts_post_thresh	149485

fold_3/logs.models.fold_3.ENCSR403GBR/logfile.modelling.fold_3.ENCSR403GBR.chrombpnet_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/model_outputs/models/chrombpnet.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/new_model_format/chrombpnet.tar with get_new_tf_model_format.py

fold_3/logs.models.fold_3.ENCSR403GBR/logfile.modelling.fold_3.ENCSR403GBR.chrombpnet_model_params.tsv

@@ -0,0 +1,9 @@
+counts_loss_weight	32.7
+filters	512
+n_dil_layers	8
+bias_model_path	/scratch/users/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/model_outputs/models/bias_model_scaled.h5
+inputlen	2114
+outputlen	1000
+max_jitter	500
+chr_fold_path	/oak/stanford/groups/akundaje/projects/chromatin-atlas-2022/splits/fold_3.json
+negative_sampling_ratio	0.1

fold_3/logs.models.fold_3.ENCSR403GBR/logfile.modelling.fold_3.ENCSR403GBR.chrombpnet_no_bias_formatting.stdout.txt

@@ -0,0 +1 @@
+Converting /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/model_outputs/models/chrombpnet_nobias.h5 to /oak/stanford/groups/akundaje/vhecht/chromatin-atlas-2022/ATAC/ENCSR403GBR/fold_3/new_model_format/chrombpnet_nobias.tar with get_new_tf_model_format.py