Merge pull request #206 from SpatialHackathon/wip_wf_optargs

(WIP) Snakemake workflow update

workflows/download.smk

@@ -1,25 +1,34 @@
 import os
-
 from shared.functions import get_git_directory
 
 
+# workflow specific setting
+configfile: "example_configs/download_config.yaml"
+# listed all the available datasets here
 configfile: "path_configs/datasets.yaml"
 
 
 print("Run Download Workflow")
 
+# Attach the specific github directory here
 GIT_DIR = get_git_directory(config)
 
+# Leave only datasets
 datasets = config.pop("datasets")
 
 
+# Get all the dataset folder
 def get_all_input(wildcards):
     all_folder = []
-    for dataset in datasets:
+    for dataset in config["use_datasets"]:
         all_folder.append(config["results_dir"] + "/" + dataset)
     return all_folder
 
 
+############## starting snakemake pipelines ##################
+
+
+# Defining all output wanted from this snakemake
 rule all:
     input:
         get_all_input,
@@ -29,8 +38,8 @@ rule download:
     output:
         dir=directory(config["results_dir"] + "/{dataset}"),
     conda:
-        lambda wildcards: GIT_DIR + "/" + datasets[wildcards.dataset]["env"]
+        lambda wildcards: GIT_DIR + datasets[wildcards.dataset]["env"]
     params:
-        script=lambda wildcards: GIT_DIR + "/" + datasets[wildcards.dataset]["script"],
+        script=lambda wildcards: GIT_DIR + datasets[wildcards.dataset]["script"],
     shell:
         "{params.script} -o {output.dir}"

workflows/example_configs/download_config.yaml

@@ -1,2 +1,8 @@
-git_dir: /home/ubuntu/workspace/SpaceHack2023
-results_dir: /home/ubuntu/tmp_data
+# Github repo path, modify based on your env
+git_dir: /home/jovyan/scratch/SpaceHack2/userfolders/jsun/workflow/SpaceHack2023
+# General data folder. All datasets folder will be stored here
+results_dir: /home/jovyan/scratch/SpaceHack2/userfolders/jsun/workflow
+
+# datasets to be downloaded. add/delete based on your need
+use_datasets:
+  - "libd_dlpfc"

workflows/example_configs/methods_config.yaml

@@ -1,62 +1,29 @@
-data_dir: "/home/ubuntu/tmp_data/libd_dlpfc"
-git_dir: /home/ubuntu/workspace/SpaceHack2023
-seed: 42
-technology: "Visium"
+# Github repo path, modify based on your env
+git_dir: /home/jovyan/scratch/SpaceHack2/userfolders/jsun/workflow/SpaceHack2023
+# Dataset path, modify based on your env
+data_dir: /home/jovyan/scratch/SpaceHack2/userfolders/jsun/workflow/libd_dlpfc
+
+seed: 2023
+
+# Methods to run for the pipeline, modify based on your need
 use_methods:
-  - "STAGATE"
-config_files:
-  spaGCN:
-    config_1: "config/config_1.json"
-    config_2: "config/config_2.json"
-    config_3: "config/config_3.json"
-    config_4: "config/config_4.json"
-    config_5: "config/config_5.json"
-    config_6: "config/config_6.json"
-    config_7: "config/config_7.json"
-    config_8: "config/config_8.json"
-    config_9: "config/config_9.json"
-    config_10: "config/config_10.json"
-    config_11: "config/config_11.json"
-    config_12: "config/config_12.json"
-    config_13: "config/config_13.json"
-    config_14: "config/config_14.json"
-    config_15: "config/config_15.json"
-    config_16: "config/config_16.json"
-  GraphST:
-    config_1: "config/config_1.json"
-    config_2: "config/config_2.json"
-    config_3: "config/config_3.json"
-    config_4: "config/config_4.json"
-    config_5: "config/config_5.json"
-    config_6: "config/config_6.json"
-  BANKSY:
-    config_1: "config/config_1.json"
-  meringue:
-    config_1: "config/config_1.json"
-  SCAN_IT:
-    config_1: "config.json"
-  scanpy:
-    config_1: "configs/config_1.json"
-  SpaceFlow:
-    config_1: "config/config.json"
-  SOTIP:
-    config_1: "config/config.json"
-  STAGATE:
-    config_1: "config/config_1.json"
-    config_2: "config/config_2.json"
-    config_3: "config/config_3.json"
-    config_4: "config/config_4.json"
-    config_5: "config/config_5.json"
-    config_6: "config/config_6.json"
-    config_7: "config/config_7.json"
-    config_8: "config/config_8.json"
-    config_9: "config/config_9.json"
-    config_10: "config/config_10.json"
-    config_11: "config/config_11.json"
-    config_12: "config/config_12.json"
-    config_13: "config/config_13.json"
-    config_14: "config/config_14.json"
-    config_15: "config/config_15.json"
-    config_16: "config/config_16.json"
-    config_17: "config/config_17.json"
-    config_18: "config/config_18.json"
+#  - "bass"
+  - "BayesSpace"
+#  - "DRSC"
+#  - "GraphST"
+#  - "SEDR"
+#  - "SOTIP"
+#  - "SpiceMix"     # GPU access
+#  - "maple"
+#  - "precast"
+#  - "SC_MEB"
+#  - "spaGCN"
+#  - "stardust"
+#  - "DeepST"
+#  - "STAGATE"       # res not n_clust
+#  - "scanpy"
+#  - "SpaceFlow"
+#  - "seurat"        # config 5 TODO
+#  - "BANKSY"
+#  - "SCAN-IT"       # buggy:data shape TODO
+#  - "meringue"      # buggy:no library found??? TODO

workflows/example_configs/metrics_config.yaml

@@ -1,2 +1,22 @@
-data_dir: "/home/ubuntu/tmp_data/libd_dlpfc"
-git_dir: /home/ubuntu/workspace/SpaceHack2023
+# Github repo path, modify based on your env
+git_dir: /home/jovyan/scratch/SpaceHack2/userfolders/jsun/workflow/SpaceHack2023
+# Dataset path, modify based on your env
+data_dir: /home/jovyan/scratch/SpaceHack2/userfolders/jsun/workflow/libd_dlpfc
+
+use_metrics:
+  - "ARI"
+#  - "Completeness"
+#  - "Entropy"
+#  - "FMI"
+#  - "Homogeneity"
+#  - "MCC"
+#  - "NMI"
+#  - "domain-specific-f1"
+#  - "jaccard"
+#  - "V_measure"                        #Config, GT
+#  - "LISI"                             #Config, embed, GT
+#  - "cluster-specific-silhouette"      #embed, no GT
+#  - "Calinski-Harabasz"
+#  - "Davies-Bouldin"
+#  - "CHAOS"                            #phyiscal coord only
+#  - "PAS"      # buggy TODO

workflows/example_configs/preprocessing_config.yaml

@@ -1,3 +1,6 @@
-data_dir: "/home/ubuntu/tmp_data/libd_dlpfc"
-git_dir: /home/ubuntu/workspace/SpaceHack2023
+# Github repo path, modify based on your env
+git_dir: /home/jovyan/scratch/SpaceHack2/userfolders/jsun/workflow/SpaceHack2023
+# Dataset path, modify based on your env
+data_dir: /home/jovyan/scratch/SpaceHack2/userfolders/jsun/workflow/libd_dlpfc
+
 n_pcs: "20"