From 474f94d7c117e0362b48815535198da4f4d4d85e Mon Sep 17 00:00:00 2001
From: Artur Szalata <arturs6668@gmail.com>
Date: Thu, 17 Oct 2024 20:33:43 +0200
Subject: [PATCH] add cpa

---
 src/methods/cpa/config.vsh.yaml | 78 ++++++++++-----------------------
 1 file changed, 22 insertions(+), 56 deletions(-)

diff --git a/src/methods/cpa/config.vsh.yaml b/src/methods/cpa/config.vsh.yaml
index 00b46c3f..6086e668 100644
--- a/src/methods/cpa/config.vsh.yaml
+++ b/src/methods/cpa/config.vsh.yaml
@@ -1,76 +1,42 @@
-# The API specifies which type of component this is.
-# It contains specifications for:
-#   - The input/output files
-#   - Common parameters
-#   - A unit test
 __merge__: ../../api/comp_method.yaml
 
-# A unique identifier for your component (required).
-# Can contain only lowercase letters or underscores.
 name: cpa
-# A relatively short label, used when rendering visualisations (required)
-label: Cpa
-# A one sentence summary of how this method works (required). Used when 
-# rendering summary tables.
-summary: "FILL IN: A one sentence summary of this method."
-# A multi-line description of how this component works (required). Used
-# when rendering reference documentation.
+label: CPA
+summary: "Compositional Perturbation Autoencoder for single-cell drug response prediction"
 description: |
-  FILL IN: A (multi-line) description of how this method works.
-# references:
-#   doi: 
-#     - 10.1000/xx.123456.789
-#   bibtex:
-#     - |
-#       @article{foo,
-#         title={Foo},
-#         author={Bar},
-#         journal={Baz},
-#         year={2024}
-#       }
-links:
-  # URL to the documentation for this method (required).
-  documentation: https://url.to/the/documentation
-  # URL to the code repository for this method (required).
-  repository: https://github.com/organisation/repository
+  CPA (Compositional Perturbation Autoencoder) is a deep learning model designed for predicting 
+  single-cell drug responses. It uses an autoencoder architecture with adversarial training to 
+  learn drug-induced changes in gene expression across different cell types and conditions. 
+  The model is trained on single-cell RNA-seq data and can predict cellular responses to 
+  unseen drug combinations and doses.
 
+  The implementation is based on https://github.com/theislab/cpa?tab=readme-ov-file and https://colab.research.google.com/github/theislab/cpa/blob/master/docs/tutorials/combosciplex_Rdkit_embeddings.ipynb#scrollTo=f46e39ed
 
+links:
+  repository: https://github.com/theislab/cpa?tab=readme-ov-file
 
-# Metadata for your component
 info:
-  # Which normalisation method this component prefers to use (required).
-  preferred_normalization: log_cp10k
-
-# Component-specific parameters (optional)
-# arguments:
-#   - name: "--n_neighbors"
-#     type: "integer"
-#     default: 5
-#     description: Number of neighbors to use.
+  # we normalize in the script
+  preferred_normalization: raw_counts
 
-# Resources required to run the component
 resources:
-  # The script of your component (required)
   - type: python_script
     path: script.py
-  # Additional resources your script needs (optional)
-  # - type: file
-  #   path: weights.pt
 
 engines:
-  # Specifications for the Docker image for this component.
   - type: docker
-    image: openproblems/base_python:1.0.0
-    # Add custom dependencies here (optional). For more information, see
-    # https://viash.io/reference/config/engines/docker/#setup .
-    # setup:
-    #   - type: python
-    #     packages: numpy<2
+    image: openproblems/base_pytorch_nvidia:1.0.0
+    setup:
+      - type: python
+        packages:
+          - cpa
+          - scanpy
+          - anndata
+          - numpy
+          - pandas
 
 runners:
-  # This platform allows running the component natively
   - type: executable
-  # Allows turning the component into a Nextflow module / pipeline.
   - type: nextflow
     directives:
-      label: [midtime,midmem,midcpu]
+      label: [ hightime, highmem, highcpu, gpu ]
\ No newline at end of file