Upload README.md with huggingface_hub

README.md

@@ -1,27 +1,81 @@
----
-dataset_info:
-  features:
-  - name: A
-    dtype: string
-  - name: B
-    dtype: string
-  - name: SeqA
-    dtype: string
-  - name: SeqB
-    dtype: string
-  - name: OrgA
-    dtype: string
-  - name: OrgB
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 775478542
-    num_examples: 530601
-  download_size: 443888825
-  dataset_size: 775478542
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
----
+---
+dataset_info:
+  features:
+  - name: A
+    dtype: string
+  - name: B
+    dtype: string
+  - name: SeqA
+    dtype: string
+  - name: SeqB
+    dtype: string
+  - name: OrgA
+    dtype: string
+  - name: OrgB
+    dtype: string
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+---
+
+# BIOGRID release 5.0.253
+
+This dataset is generated from BioGRID MV using `Synthyra/BIOGRID-MV-5.0.253`.
+BioGRID downloads are hosted here: https://downloads.thebiogrid.org/Download/BioGRID/Release-Archive/BIOGRID-5.0.253/
+The exact zip used for this split: https://downloads.thebiogrid.org/Download/BioGRID/Release-Archive/BIOGRID-5.0.253/BIOGRID-MV-Physical-5.0.253.tab3.zip
+
+Swiss-Prot and TrEMBL accessions are resolved to current UniProtKB primary accessions using the UniProt ID Mapping service: https://www.uniprot.org/id-mapping
+We keep interaction pairs only when both partners resolve to a UniProtKB sequence.
+When multiple accessions are present, Swiss-Prot IDs are preferred over TrEMBL.
+
+## Processing Overview
+- Download and unzip BioGRID
+- Stream BioGRID tab3 rows in chunks
+- Normalize accessions (optionally strip isoform suffixes)
+- Resolve accessions to UniProtKB primary IDs and sequences
+- Emit rows with A/B, SeqA/SeqB, OrgA/OrgB
+- Push dataset shards to the Hugging Face Hub
+
+## Key Script Snippets
+```python
+# Download + unzip
+zip_path = _ensure_biogrid_zip(version, kind=kind, data_dir=data_dir)
+
+# Stream BioGRID rows from the tab3 file inside the zip
+for chunk in _iter_biogrid_chunks(zip_path, chunksize=chunksize):
+    ...
+
+# Map accessions to UniProt primary + fetch sequences
+id_to_primary, primary_to_seq = _resolve_sequences_for_ids(...)
+
+# Build rows and push to Hub
+dataset = Dataset.from_generator(_row_generator, gen_kwargs={...}, features=features)
+dataset.push_to_hub(repo_id, token=token, private=args.private, max_shard_size=args.max_shard_size)
+```
+
+
+This dataset is the Multi-Validated (MV) version of BioGRID. The selection criteria is summarized in the image below.
+
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/62f2bd3bdb7cbd214b658c48/8f-34ZCfGxamRvQxmqHwL.png)
+
+## Citation
+
+If you use this dataset please cite their work:
+```
+@article{Oughtred2020,
+  title = {The <scp>BioGRID</scp> database: A comprehensive biomedical resource of curated protein,  genetic,  and chemical interactions},
+  volume = {30},
+  ISSN = {1469-896X},
+  url = {http://dx.doi.org/10.1002/pro.3978},
+  DOI = {10.1002/pro.3978},
+  number = {1},
+  journal = {Protein Science},
+  publisher = {Wiley},
+  author = {Oughtred,  Rose and Rust,  Jennifer and Chang,  Christie and Breitkreutz,  Bobby‐Joe and Stark,  Chris and Willems,  Andrew and Boucher,  Lorrie and Leung,  Genie and Kolas,  Nadine and Zhang,  Frederick and Dolma,  Sonam and Coulombe‐Huntington,  Jasmin and Chatr‐aryamontri,  Andrew and Dolinski,  Kara and Tyers,  Mike},
+  year = {2020},
+  month = nov,
+  pages = {187–200}
+}
+```