norm for grnboost

openproblems-bio · Sep 17, 2024 · 3dd3948 · 3dd3948
1 parent ada5099
commit 3dd3948
Show file tree

Hide file tree

Showing 10 changed files with 1,237 additions and 1,059 deletions.
diff --git a/dockerfiles/scenic/Dockerfile b/dockerfiles/scenic/Dockerfile
@@ -0,0 +1,11 @@
+FROM python:3.6.8-slim
+
+RUN BUILDPKGS="build-essential apt-utils" && \
+    apt-get update && \
+    apt-get install -y $BUILDPKGS && \
+    apt-get install -y procps && \
+    rm -rf /var/lib/apt/lists/*
+
+RUN pip install --no-cache-dir --upgrade pyscenic==0.9.6 dask==1.0.0 pandas==0.23.4 scanpy==
+
+RUN apt-get remove --purge -y $BUILDPKGS
diff --git a/runs.ipynb b/runs.ipynb
diff --git a/scripts/run_grn_evaluation.sh b/scripts/run_grn_evaluation.sh
@@ -3,13 +3,13 @@
 # RUN_ID="run_$(date +%Y-%m-%d_%H-%M-%S)"
 # reg_type=${1} #GB, ridge
 # viash ns build --parallel
-reg_type=ridge
+reg_type=${2}
 
-RUN_ID="celltype_donor_0_subset_${reg_type}"
+RUN_ID=${1}_${2}
 resources_dir="s3://openproblems-data/resources/grn"
 # resources_dir="./resources"
 publish_dir="${resources_dir}/results/${RUN_ID}"
-grn_models_folder="${resources_dir}/grn_models/donor_0_celltype"
+grn_models_folder="${resources_dir}/grn_models/alldonors_default"
 
 subsample=-2
 max_workers=10
@@ -19,18 +19,19 @@ metric_ids="[regression_1, regression_2]"
 param_file="./params/${RUN_ID}.yaml"
 
 grn_names=(
-    # "scglue"
-    # "scenicplus"
-    # "celloracle"
-    # "granie"
-    # "figr"
-    # "collectri"
-    # "genie3"
+    "scglue"
+    "scenicplus"
+    "celloracle"
+    "granie"
+    "figr"
+    "collectri"
+    "genie3"
     "grnboost2"
-    # "ppcor"
+    "ppcor"
     "portia"
     "positive_control"
     "pearson_causal"
+    "pearson"
     )
 
 # Start writing to the YAML file

diff --git a/src/api/comp_method.yaml b/src/api/comp_method.yaml
@@ -41,7 +41,7 @@ functionality:
     - name: --cell_type_specific
       type: boolean
       direction: input
-      default: true
+      default: false
 
 
 

diff --git a/src/control_methods/baseline_corr/script.py b/src/control_methods/baseline_corr/script.py
@@ -9,15 +9,16 @@
 
 ## VIASH START
 par = {
-    'multiomics_rna': 'resources/grn-benchmark/perturbation_data.h5ad',
+    'multiomics_rna': 'resources/grn-benchmark/multiomics_rna.h5ad',
     'tf_all': 'resources/prior/tf_all.csv',
-    'causal': True,
+    'causal': False,
     'metacell': False,
-    'cell_type_specific': True,
+    'cell_type_specific': False,
     'impute': False,
     'max_n_links': 50000,
     'corr_method': 'pearson',
-    'prediction': 'resources/grn_models/positive_control.csv'
+    'prediction': 'resources/grn_models/alldonors_default/pearson.csv',
+    "seed": 32
 }
 ## VIASH END
 print(par)

diff --git a/src/methods/multi_omics/granie/run.sh b/src/methods/multi_omics/granie/run.sh
@@ -0,0 +1,3 @@
+viash run src/methods/granie/config.vsh.yaml -- --multiomics_rna resources_test/grn-benchmark/multiomics_rna.h5ad \
+    --multiomics_atac resources_test/grn-benchmark/multiomics_atac.h5ad \
+    --prediction output/prediction.csv