Upload BPNet SOX2 ChIP-seq model (exp26_SOX2)

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+exp26_SOX2.torch filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,111 @@
+---
+license: mit
+tags:
+  - bpnet
+  - bpnet-lite
+  - genomics
+  - chip-seq
+  - transcription-factor
+  - sox2
+  - mm10
+library_name: bpnet-lite
+---
+
+# BPNet_ChIP-seq_SOX2
+
+A [BPNet](https://www.nature.com/articles/s41588-021-00782-6) model trained with
+[`bpnet-lite`](https://github.com/jmschrei/bpnet-lite) on SOX2 ChIP-seq data,
+used to model base-resolution binding signal and to discover sequence motifs via
+TF-MoDISco.
+
+## Model details
+
+- **Architecture:** BPNet (`bpnet-lite` implementation), ? filters,
+  ? dilated convolutional layers, with bias tracks for both the
+  profile and total-count heads.
+- **Genome:** mm10
+- **Input window:** 2114 bp sequence -> **output window:**
+  1000 bp profile prediction
+- **Training data:** SOX2 ChIP-seq, MACS2 narrow peaks as positive loci,
+  with a matched negative set.
+  - Peaks (loci): `/scratch/prj/stem_cells_pituitary/Bence/atacseq_chip_results/26/bwa/merged_library/macs2/narrow_peak/SOX2_REP1.mLb.clN_peaks.bed`
+  - Signal (BAM): `exp26_SOX2.bw`
+  - Negatives: `/scratch/prj/stem_cells_pituitary/Bence/atacseq_chip_results/26/bwa/merged_library/macs2/narrow_peak/SOX2_negatives_bpnet.bed`
+- **Train/validation split (by chromosome):**
+  - Validation: ``
+  - Training: all other chromosomes (see `exp26_SOX2.bpnet.fit.json` for the
+    exact list)
+- **Training hyperparameters:** lr=None, batch_size=None,
+  max_epochs=None, early_stopping=None epochs,
+  max_jitter=None, reverse_complement_augmentation=None
+- **Checkpoint selection:** `exp26_SOX2.torch` is the best-validation-loss checkpoint saved
+  during training by `bpnet-lite`'s `fit` routine (as opposed to a `.final.torch` checkpoint,
+  which reflects the last training epoch and is not included here).
+
+## Performance
+
+Reported metrics from `exp26_SOX2.performance.tsv`:
+
+```
+profile_mnll	profile_jsd	profile_pearson	profile_spearman	count_pearson	count_spearman	count_mse
+472.9859924316406	0.44859686493873596	0.20588907599449158	0.06264603137969971	0.39160212874412537	0.3859022855758667	0.6483629941940308
+```
+
+See `exp26_SOX2.evaluate.json` for the full evaluation output.
+
+## How to load
+
+This model was trained with `bpnet-lite` and is saved as a raw PyTorch `state_dict`-style
+`.torch` checkpoint (not a `transformers`-compatible format). To load it, install
+`bpnet-lite` and use its `BPNet` class directly:
+
+```bash
+pip install bpnet-lite
+```
+
+```python
+import torch
+from bpnetlite import BPNet
+
+# Re-create the architecture with the same hyperparameters used in training
+# (see exp26_SOX2.bpnet.fit.json for the full config)
+model = BPNet(
+    n_filters=64,
+    n_layers=8,
+    profile_output_bias=True,
+    count_output_bias=True,
+)
+
+state_dict = torch.load("exp26_SOX2.torch", map_location="cpu")
+model.load_state_dict(state_dict)
+model.eval()
+```
+
+> Double check the exact `BPNet(...)` constructor signature against the `bpnet-lite`
+> version you have installed, since argument names/defaults can change between releases.
+> Pin the same `bpnet-lite` version that produced this model if you need exact
+> reproducibility — see `exp26_SOX2.log` for environment details if recorded.
+
+## Files
+
+| File | Description |
+|---|---|
+| `exp26_SOX2.torch` | Model weights (best validation checkpoint) |
+| `exp26_SOX2.bpnet.fit.json` | Full training configuration (architecture, data paths, chrom split, hyperparameters) |
+| `exp26_SOX2.bpnet.attribute.json` | Configuration used for computing attributions downstream |
+| `exp26_SOX2.performance.tsv` | Reported performance metrics |
+| `exp26_SOX2.evaluate.json` | Full evaluation output |
+| `exp26_SOX2.log` | Training log |
+
+## Citation
+
+If you use this model, please cite the original BPNet paper and the `bpnet-lite`
+software:
+
+- Avsec et al., "Base-resolution models of transcription-factor binding reveal soft
+  motif syntax", *Nature Genetics*, 2021.
+- `bpnet-lite`: https://github.com/jmschrei/bpnet-lite
+
+## Repo
+
+`ge8rgia/BPNet_ChIP-seq_SOX2`

exp26_SOX2.bpnet.attribute.json

@@ -0,0 +1,25 @@
+{
+    "attr_filename": "exp26_SOX2.attributions.attr.npz",
+    "batch_size": 64,
+    "chroms": [
+        "chr8",
+        "chr20"
+    ],
+    "device": "cuda",
+    "exclusion_lists": null,
+    "idx_filename": "exp26_SOX2.attributions.idxs.npy",
+    "in_window": 2114,
+    "loci": [
+        "/scratch/prj/stem_cells_pituitary/Bence/atacseq_chip_results/26/bwa/merged_library/macs2/narrow_peak/SOX2_REP1.mLb.clN_peaks.bed"
+    ],
+    "model": "exp26_SOX2.torch",
+    "n_shuffles": 20,
+    "ohe_filename": "exp26_SOX2.attributions.ohe.npz",
+    "out_window": 1000,
+    "output": "counts",
+    "random_state": null,
+    "sequences": "/scratch/prj/stem_cells_pituitary/Georgia/genome/mm10/mm10.fa",
+    "skip": false,
+    "verbose": true,
+    "warning_threshold": 0.001
+}

exp26_SOX2.bpnet.fit.json

@@ -0,0 +1,64 @@
+{
+    "batch_size": 64,
+    "controls": null,
+    "count_output_bias": true,
+    "device": "cuda",
+    "dtype": "float32",
+    "early_stopping": 5,
+    "exclusion_lists": null,
+    "in_window": 2114,
+    "loci": [
+        "/scratch/prj/stem_cells_pituitary/Bence/atacseq_chip_results/26/bwa/merged_library/macs2/narrow_peak/SOX2_REP1.mLb.clN_peaks.bed"
+    ],
+    "lr": 0.001,
+    "max_counts": 99999999,
+    "max_epochs": 20,
+    "max_jitter": 128,
+    "min_counts": 0,
+    "n_filters": 64,
+    "n_layers": 8,
+    "name": "exp26_SOX2",
+    "negative_ratio": 0.333,
+    "negatives": [
+        "/scratch/prj/stem_cells_pituitary/Bence/atacseq_chip_results/26/bwa/merged_library/macs2/narrow_peak/SOX2_negatives_bpnet.bed"
+    ],
+    "out_window": 1000,
+    "performance_filename": "exp26_SOX2.performance.tsv",
+    "profile_output_bias": true,
+    "random_state": null,
+    "reverse_complement": true,
+    "reverse_complement_average": true,
+    "scheduler": true,
+    "sequences": "/scratch/prj/stem_cells_pituitary/Georgia/genome/mm10/mm10.fa",
+    "signals": [
+        "exp26_SOX2.bw"
+    ],
+    "skip": false,
+    "summits": false,
+    "training_chroms": [
+        "chr2",
+        "chr4",
+        "chr5",
+        "chr7",
+        "chr9",
+        "chr10",
+        "chr11",
+        "chr12",
+        "chr13",
+        "chr14",
+        "chr15",
+        "chr16",
+        "chr17",
+        "chr18",
+        "chr19",
+        "chr21",
+        "chr22",
+        "chrX",
+        "chrY"
+    ],
+    "validation_chroms": [
+        "chr8",
+        "chr20"
+    ],
+    "verbose": true
+}

exp26_SOX2.evaluate.json

@@ -0,0 +1,70 @@
+{
+    "batch_size": 64,
+    "chroms": [
+        "chr8",
+        "chr20"
+    ],
+    "controls": null,
+    "count_loss_weight": 236.43389892578125,
+    "count_output_bias": true,
+    "device": "cuda",
+    "dtype": "float32",
+    "early_stopping": 5,
+    "exclusion_lists": null,
+    "in_window": 2114,
+    "loci": [
+        "/scratch/prj/stem_cells_pituitary/Bence/atacseq_chip_results/26/bwa/merged_library/macs2/narrow_peak/SOX2_REP1.mLb.clN_peaks.bed"
+    ],
+    "lr": 0.001,
+    "max_counts": 99999999,
+    "max_epochs": 20,
+    "max_jitter": 0,
+    "min_counts": 0,
+    "model": "exp26_SOX2.torch",
+    "n_filters": 64,
+    "n_layers": 8,
+    "name": "exp26_SOX2",
+    "negative_ratio": 0.333,
+    "negatives": [
+        "/scratch/prj/stem_cells_pituitary/Bence/atacseq_chip_results/26/bwa/merged_library/macs2/narrow_peak/SOX2_negatives_bpnet.bed"
+    ],
+    "out_window": 1000,
+    "performance_filename": "exp26_SOX2.performance.tsv",
+    "profile_output_bias": true,
+    "random_state": null,
+    "reverse_complement": false,
+    "reverse_complement_average": true,
+    "scheduler": true,
+    "sequences": "/scratch/prj/stem_cells_pituitary/Georgia/genome/mm10/mm10.fa",
+    "signals": [
+        "exp26_SOX2.bw"
+    ],
+    "skip": false,
+    "summits": false,
+    "training_chroms": [
+        "chr2",
+        "chr4",
+        "chr5",
+        "chr7",
+        "chr9",
+        "chr10",
+        "chr11",
+        "chr12",
+        "chr13",
+        "chr14",
+        "chr15",
+        "chr16",
+        "chr17",
+        "chr18",
+        "chr19",
+        "chr21",
+        "chr22",
+        "chrX",
+        "chrY"
+    ],
+    "validation_chroms": [
+        "chr8",
+        "chr20"
+    ],
+    "verbose": true
+}

exp26_SOX2.log

@@ -0,0 +1,17 @@
+Epoch	Iteration	Training Time	Validation Time	Training MNLL	Training Count MSE	Validation MNLL	Validation Profile Pearson	Validation Count Pearson	Validation Count MSE	Saved?
+0	1081	12.538955688476562	0.22972440719604492	493.6463928222656	1.5571314096450806	498.90155029296875	0.1414016	0.25350755	0.9193392395973206	True
+1	2162	11.227972507476807	0.18920421600341797	468.42889404296875	1.212290644645691	489.6493835449219	0.16447474	0.2962626	0.794971227645874	True
+2	3243	11.230249166488647	0.18810677528381348	501.5304260253906	1.641969084739685	483.093017578125	0.17985299	0.3210969	0.787030041217804	True
+3	4324	11.219478607177734	0.18841266632080078	388.6416320800781	1.202521800994873	479.6177062988281	0.18650699	0.349209	0.7325787544250488	True
+4	5405	11.22038745880127	0.1877729892730713	387.3236083984375	0.5208539962768555	477.2947082519531	0.19270068	0.35677594	0.8444055318832397	False
+5	6486	11.229341745376587	0.18797850608825684	428.5082092285156	1.042463779449463	476.99267578125	0.19493777	0.36274323	0.722731351852417	True
+6	7567	11.225963115692139	0.18822121620178223	476.11187744140625	1.4584213495254517	476.9776306152344	0.19372638	0.3682373	0.9221944212913513	False
+7	8648	11.240039825439453	0.1881241798400879	502.65234375	1.389481544494629	476.1083679199219	0.19767307	0.37159348	0.8530063629150391	False
+8	9729	11.23106074333191	0.18815875053405762	537.5493774414062	1.5133410692214966	475.6878662109375	0.19802807	0.37246436	0.9266992807388306	False
+9	10810	11.224467277526855	0.18801069259643555	475.5256652832031	1.4387134313583374	476.7750549316406	0.19580016	0.38152167	1.542570948600769	False
+10	11891	11.22986125946045	0.188140869140625	449.2487487792969	0.7869852185249329	475.4999694824219	0.19919328	0.372169	0.6735904216766357	True
+11	12972	11.223177909851074	0.18834972381591797	479.9490051269531	0.8299573659896851	473.94085693359375	0.20223816	0.386478	0.8706695437431335	False
+12	14053	11.227935314178467	0.1881556510925293	359.5709533691406	0.7716255187988281	474.9963073730469	0.20094658	0.38978532	0.752129852771759	False
+13	15134	11.231069326400757	0.18865060806274414	456.62451171875	0.6766591668128967	473.44091796875	0.20359589	0.39161077	0.8789016008377075	False
+14	16215	11.22939133644104	0.18801569938659668	474.4501647949219	0.7873633503913879	476.0885009765625	0.19960625	0.3962401	0.7771286368370056	False
+15	17296	11.224156856536865	0.1884617805480957	486.74066162109375	0.7792139053344727	473.2563171386719	0.2054644	0.38726386	0.7868853211402893	False

exp26_SOX2.performance.tsv

@@ -0,0 +1,2 @@
+profile_mnll	profile_jsd	profile_pearson	profile_spearman	count_pearson	count_spearman	count_mse
+472.9859924316406	0.44859686493873596	0.20588907599449158	0.06264603137969971	0.39160212874412537	0.3859022855758667	0.6483629941940308

exp26_SOX2.torch

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d420aa2b608d4167dab8ca4c2b3b0bef8e235c7369ced51f8a62dd10f0b069bd
+size 453173