Add mlflow models

CHANGELOG.md

@@ -3,6 +3,7 @@
 ## NEW FUNCTIONALITY
 
 * Add new method: CellMapper, which is a k-NN based approach to map cells across representations and can be used for label projection. Two versions are included here, one based on PCA or CCA embeddings (`linear`) and one based on an scvi embedding (`scvi`) (PR #22)
+* Add MLflow-based methods: Geneformer, scGPT, scVI, TranscriptFormer, and UCE for label projection using pre-trained foundation models (PR #28)
 
 ## MAJOR CHANGES
 

src/methods/geneformer_mlflow/config.vsh.yaml

@@ -0,0 +1,50 @@
+__merge__: ../../api/base_method.yaml
+
+name: geneformer_mlflow
+label: Geneformer (MLflow)
+summary: Geneformer is a foundational transformer model pretrained on a large-scale corpus of single cell transcriptomes to enable context-aware predictions in settings with limited data in network biology (MLflow)
+description: |
+  Geneformer is a context-aware, attention-based deep learning model pretrained
+  on a large-scale corpus of single-cell transcriptomes to enable
+  context-specific predictions in settings with limited data in network biology.
+
+  This version uses a pre-trained MLflow model. A kNN classifier is trained on
+  embeddings for the training data and used to predict labels for the test
+  data. It does not use the built-in Geneformer classifier.
+references:
+  doi:
+    - 10.1038/s41586-023-06139-9
+    - 10.1101/2024.08.16.608180
+links:
+  documentation: https://geneformer.readthedocs.io/en/latest/index.html
+  repository: https://huggingface.co/ctheodoris/Geneformer
+
+info:
+  preferred_normalization: counts
+
+arguments:
+  - name: --model
+    type: file
+    description: |
+      An MLflow model URI for the Geneformer model. If it is a .zip or
+      .tar.gz file it will be extracted to a temporary directory.
+    required: true
+
+resources:
+  - type: python_script
+    path: script.py
+  - path: /src/utils/exit_codes.py
+  - path: /src/utils/unpack.py
+  - path: /src/utils/mlflow.py
+  - path: requirements.txt
+
+engines:
+  - type: docker
+    image: openproblems/base_pytorch_nvidia:1
+    __merge__: /src/utils/mlflow_docker_setup.yaml
+
+runners:
+  - type: executable
+  - type: nextflow
+    directives:
+      label: [hightime, highmem, midcpu, gpu]