Evo into Release

.github/workflows/main.yml

@@ -152,6 +152,7 @@ jobs:
           sed -i 's/get_anndata_from_hf_dataset(ds\[\\"test\\"\])/get_anndata_from_hf_dataset(ds\[\\"test\\"\])[:10]/g' ./examples/notebooks/Cell-Type-Classification-Fine-Tuning.ipynb
           sed -i 's/list(np.array(train_dataset.obs\[\\"LVL1\\"].tolist()))/list(np.array(train_dataset.obs\[\\"LVL1\\"].tolist()))[:100]/g' ./examples/notebooks/Cell-Type-Classification-Fine-Tuning.ipynb
           sed -i 's/list(np.array(test_dataset.obs\[\\"LVL1\\"].tolist()))/list(np.array(test_dataset.obs\[\\"LVL1\\"].tolist()))[:10]/g' ./examples/notebooks/Cell-Type-Classification-Fine-Tuning.ipynb
+          rm ./examples/notebooks/Evo-2.ipynb
 
       - name: Run Notebooks
         run: |

.github/workflows/release.yml

@@ -170,6 +170,7 @@ jobs:
           sed -i 's/get_anndata_from_hf_dataset(ds\[\\"test\\"\])/get_anndata_from_hf_dataset(ds\[\\"test\\"\])[:10]/g' ./examples/notebooks/Cell-Type-Classification-Fine-Tuning.ipynb
           sed -i 's/list(np.array(train_dataset.obs\[\\"LVL1\\"].tolist()))/list(np.array(train_dataset.obs\[\\"LVL1\\"].tolist()))[:100]/g' ./examples/notebooks/Cell-Type-Classification-Fine-Tuning.ipynb
           sed -i 's/list(np.array(test_dataset.obs\[\\"LVL1\\"].tolist()))/list(np.array(test_dataset.obs\[\\"LVL1\\"].tolist()))[:10]/g' ./examples/notebooks/Cell-Type-Classification-Fine-Tuning.ipynb
+          rm ./examples/notebooks/Evo-2.ipynb
 
       - name: Run Notebooks
         run: |

.gitignore

@@ -28,6 +28,7 @@ results*
 
 # Log
 debug.log
+*debug.log
 
 # Mac Files
 *.DS_Store

README.md

@@ -30,6 +30,10 @@ Let’s build the most exciting AI-for-Bio community together!
 </div>
 
 ## What's new?
+### Evo2
+We have integrated [Evo2](https://github.com/ArcInstitute/evo2) into our helical package and have made a model card for it in our [Evo2 model folder](helical/helical/models/evo_2/README.md). If you would like to test the model, take a look at our [example notebook](helical/examples/notebooks/Evo-2.ipynb)!
+Let us know what you think and we are happy to help you with the larger model (40B parameters!) if needed!
+
 ### 🧬 Introducing Helix-mRNA-v0: Unlocking new frontiers & use cases in mRNA therapy 🧬
 We’re thrilled to announce the release of our first-ever mRNA Bio Foundation Model, designed to:
 
@@ -151,6 +155,7 @@ A lot of our models have been published by talend authors developing these excit
 - [scikit-learn](https://github.com/scikit-learn/scikit-learn)
 - [GenePT](https://github.com/yiqunchen/GenePT)
 - [Caduceus](https://github.com/kuleshov-group/caduceus)
+- [Evo2](https://github.com/ArcInstitute/evo2)
 
 ### Licenses
 
@@ -162,6 +167,7 @@ You can find the Licenses for each model implementation in the model repositorie
 - [Geneformer](https://github.com/helicalAI/helical/blob/release/helical/models/geneformer/LICENSE)
 - [UCE](https://github.com/helicalAI/helical/blob/release/helical/models/uce/LICENSE)
 - [HyenaDNA](https://github.com/helicalAI/helical/blob/release/helical/models/hyena_dna/LICENSE)
+- [Evo2](https://github.com/helicalAI/helical/blob/release/helical/models/evo_2/LICENSE)
 
 ## Citation
 

ci/download_all.py

@@ -1,18 +1,20 @@
 from helical.utils.downloader import Downloader
 from pathlib import Path
 import logging
+
 LOGGER = logging.getLogger(__name__)
 
+
 def download_geneformer_models():
     downloader = Downloader()
-    versions = ['v1', 'v2']
+    versions = ["v1", "v2"]
 
     # We can decide to download more models by simply adding the model names from the full list as reported in geneformer_config.py
     version_models_dict = {
         "v1": ["gf-12L-30M-i2048", "gf-6L-30M-i2048"],
         "v2": ["gf-12L-95M-i4096", "gf-12L-95M-i4096-CLcancer", "gf-20L-95M-i4096"],
     }
-    
+
     for version in versions:
         # Download common files for each version
         common_files = [
@@ -22,31 +24,36 @@ def download_geneformer_models():
         ]
         for file in common_files:
             downloader.download_via_name(file)
-        
+
         # Get all model directories
         model_dirs = version_models_dict[version]
-        
+
         for model_name in model_dirs:
             model_files = [
                 f"geneformer/{version}/{model_name}/config.json",
                 f"geneformer/{version}/{model_name}/training_args.bin",
             ]
-            
+
             # Add version-specific files
-            if version == 'v2':
-                model_files.extend([
-                    f"geneformer/{version}/{model_name}/generation_config.json",
-                    f"geneformer/{version}/{model_name}/model.safetensors",
-                ])
+            if version == "v2":
+                model_files.extend(
+                    [
+                        f"geneformer/{version}/{model_name}/generation_config.json",
+                        f"geneformer/{version}/{model_name}/model.safetensors",
+                    ]
+                )
             else:
-                model_files.append(f"geneformer/{version}/{model_name}/pytorch_model.bin")
-
+                model_files.append(
+                    f"geneformer/{version}/{model_name}/pytorch_model.bin"
+                )
+
             # Download all files for the current model
             for file in model_files:
                 downloader.download_via_name(file)
-    
+
     LOGGER.info("All Geneformer models and files have been downloaded.")
 
+
 def main():
     downloader = Downloader()
     downloader.display = False
@@ -55,14 +62,30 @@ def main():
     downloader.download_via_name("uce/all_tokens.torch")
     downloader.download_via_name("uce/species_chrom.csv")
     downloader.download_via_name("uce/species_offsets.pkl")
-    downloader.download_via_name("uce/protein_embeddings/Homo_sapiens.GRCh38.gene_symbol_to_embedding_ESM2.pt")
-    downloader.download_via_name("uce/protein_embeddings/Macaca_fascicularis.Macaca_fascicularis_6.0.gene_symbol_to_embedding_ESM2.pt")
-    downloader.download_via_name("uce/protein_embeddings/Danio_rerio.GRCz11.gene_symbol_to_embedding_ESM2.pt")
-    downloader.download_via_name("uce/protein_embeddings/Macaca_mulatta.Mmul_10.gene_symbol_to_embedding_ESM2.pt")
-    downloader.download_via_name("uce/protein_embeddings/Microcebus_murinus.Mmur_3.0.gene_symbol_to_embedding_ESM2.pt")
-    downloader.download_via_name("uce/protein_embeddings/Mus_musculus.GRCm39.gene_symbol_to_embedding_ESM2.pt")
-    downloader.download_via_name("uce/protein_embeddings/Sus_scrofa.Sscrofa11.1.gene_symbol_to_embedding_ESM2.pt")
-    downloader.download_via_name("uce/protein_embeddings/Xenopus_tropicalis.Xenopus_tropicalis_v9.1.gene_symbol_to_embedding_ESM2.pt")
+    downloader.download_via_name(
+        "uce/protein_embeddings/Homo_sapiens.GRCh38.gene_symbol_to_embedding_ESM2.pt"
+    )
+    downloader.download_via_name(
+        "uce/protein_embeddings/Macaca_fascicularis.Macaca_fascicularis_6.0.gene_symbol_to_embedding_ESM2.pt"
+    )
+    downloader.download_via_name(
+        "uce/protein_embeddings/Danio_rerio.GRCz11.gene_symbol_to_embedding_ESM2.pt"
+    )
+    downloader.download_via_name(
+        "uce/protein_embeddings/Macaca_mulatta.Mmul_10.gene_symbol_to_embedding_ESM2.pt"
+    )
+    downloader.download_via_name(
+        "uce/protein_embeddings/Microcebus_murinus.Mmur_3.0.gene_symbol_to_embedding_ESM2.pt"
+    )
+    downloader.download_via_name(
+        "uce/protein_embeddings/Mus_musculus.GRCm39.gene_symbol_to_embedding_ESM2.pt"
+    )
+    downloader.download_via_name(
+        "uce/protein_embeddings/Sus_scrofa.Sscrofa11.1.gene_symbol_to_embedding_ESM2.pt"
+    )
+    downloader.download_via_name(
+        "uce/protein_embeddings/Xenopus_tropicalis.Xenopus_tropicalis_v9.1.gene_symbol_to_embedding_ESM2.pt"
+    )
 
     downloader.download_via_name("scgpt/scGPT_CP/vocab.json")
     downloader.download_via_name("scgpt/scGPT_CP/best_model.pt")
@@ -75,23 +98,43 @@ def main():
     downloader.download_via_name("hyena_dna/hyenadna-medium-450k-seqlen.ckpt")
     downloader.download_via_name("hyena_dna/hyenadna-large-1m-seqlen.ckpt")
 
-    downloader.download_via_name('caduceus/caduceus-ph-16L-seqlen-131k-d256/model.safetensors')
-    downloader.download_via_name('caduceus/caduceus-ph-16L-seqlen-131k-d256/config.json')
-    downloader.download_via_name('caduceus/caduceus-ph-4L-seqlen-1k-d118/model.safetensors')
-    downloader.download_via_name('caduceus/caduceus-ph-4L-seqlen-1k-d118/config.json')
-    downloader.download_via_name('caduceus/caduceus-ph-4L-seqlen-1k-d256/model.safetensors')
-    downloader.download_via_name('caduceus/caduceus-ph-4L-seqlen-1k-d256/config.json')
-    downloader.download_via_name('caduceus/caduceus-ps-16L-seqlen-131k-d256/model.safetensors')
-    downloader.download_via_name('caduceus/caduceus-ps-16L-seqlen-131k-d256/config.json')
-    downloader.download_via_name('caduceus/caduceus-ps-4L-seqlen-1k-d118/model.safetensors')
-    downloader.download_via_name('caduceus/caduceus-ps-4L-seqlen-1k-d118/config.json')
-    downloader.download_via_name('caduceus/caduceus-ps-4L-seqlen-1k-d256/model.safetensors')
-    downloader.download_via_name('caduceus/caduceus-ps-4L-seqlen-1k-d256/config.json')
-
-    downloader.download_via_name('genept/genept_embeddings/genept_embeddings.json')
-
-    downloader.download_via_link(Path("yolksac_human.h5ad"), "https://huggingface.co/datasets/helical-ai/yolksac_human/resolve/main/data/17_04_24_YolkSacRaw_F158_WE_annots.h5ad?download=true")
+    downloader.download_via_name(
+        "caduceus/caduceus-ph-16L-seqlen-131k-d256/model.safetensors"
+    )
+    downloader.download_via_name(
+        "caduceus/caduceus-ph-16L-seqlen-131k-d256/config.json"
+    )
+    downloader.download_via_name(
+        "caduceus/caduceus-ph-4L-seqlen-1k-d118/model.safetensors"
+    )
+    downloader.download_via_name("caduceus/caduceus-ph-4L-seqlen-1k-d118/config.json")
+    downloader.download_via_name(
+        "caduceus/caduceus-ph-4L-seqlen-1k-d256/model.safetensors"
+    )
+    downloader.download_via_name("caduceus/caduceus-ph-4L-seqlen-1k-d256/config.json")
+    downloader.download_via_name(
+        "caduceus/caduceus-ps-16L-seqlen-131k-d256/model.safetensors"
+    )
+    downloader.download_via_name(
+        "caduceus/caduceus-ps-16L-seqlen-131k-d256/config.json"
+    )
+    downloader.download_via_name(
+        "caduceus/caduceus-ps-4L-seqlen-1k-d118/model.safetensors"
+    )
+    downloader.download_via_name("caduceus/caduceus-ps-4L-seqlen-1k-d118/config.json")
+    downloader.download_via_name(
+        "caduceus/caduceus-ps-4L-seqlen-1k-d256/model.safetensors"
+    )
+    downloader.download_via_name("caduceus/caduceus-ps-4L-seqlen-1k-d256/config.json")
+
+    downloader.download_via_name("genept/genept_embeddings/genept_embeddings.json")
+
+    downloader.download_via_link(
+        Path("yolksac_human.h5ad"),
+        "https://huggingface.co/datasets/helical-ai/yolksac_human/resolve/main/data/17_04_24_YolkSacRaw_F158_WE_annots.h5ad?download=true",
+    )
     return True
 
+
 if __name__ == "__main__":
     main()