Rework curl data download

README.md

@@ -3,10 +3,14 @@
 # Prerequisites
 
 We assume a Unix command line workflow. The following software is required to take advantage of the pre-defined workflow:
+  * curl for downloading the data (in case you do not have curl installed, it is also available from within the container)
   * [python](https://www.python.org/download/releases/3.5.1/) 3.5.1 (higher versions might work as well) 
   * [snakemake](https://snakemake.readthedocs.io/en/stable/getting_started/installation.html) version 5.2.1 (higher versions will work as well)
   * [singularity](https://www.sylabs.io/guides/2.6/user-guide/index.html) 2.6.0 (higher versions might work as well)
   
++ CENTER-TBI account and API key, store as NEUROBOT_USR and NEUROBOT_API 
+environment variables.
+
 The entire analysis is containerized using a [docker container](https://cloud.docker.com/u/kkmann/repository/docker/kkmann/gose-6mo-imputation). 
 The container can either be used to execute scripts individually inside the container, or it can be used to run the entire 
 pre-defined snakemake workflow using the container via singularity (recommended).

Snakefile

@@ -6,7 +6,7 @@ configfile: "config.yml"
 
 
 
-rule import_neurobot_csv:
+rule download_data:
     output:
         "data/{version}/df_baseline.rds",
         "data/{version}/df_ctmri.rds",
@@ -15,7 +15,7 @@ rule import_neurobot_csv:
         "data/{version}/df_gose.rds"
     shell:
         """
-        Rscript scripts/import_neurobot_data.R data/{wildcards.version} data/{wildcards.version}
+        bash scripts/download_{wildcards.version}.sh
         """
 
 
@@ -24,7 +24,7 @@ rule import_neurobot_csv:
 
 rule prepare_data:
     input:
-        rules.import_neurobot_csv.output,
+        rules.download_data.output,
         markdown = "reports/prepare_data.Rmd"
     output:
         "output/{version}/data/df_gose.rds",
@@ -39,12 +39,6 @@ rule prepare_data:
         mv reports/figures.zip {output.figures}
         """
 
-# define corresponding target rule for ease of use
-rule data_report_v1_1:
-    input:
-        pdf     = "output/v1.1/prepare_data.pdf",
-        figures = "output/v1.1/prepare_data_figures.zip"
-
 
 
 
@@ -80,14 +74,6 @@ rule generate_validation_data:
         Rscript scripts/generate_validation_data.R output/{wildcards.version}/data {config[mi_m]} {config[folds]} {config[seed]}
         """
 
-rule generate_validation_data_v1_1:
-    input:
-        ["output/v1.1/data/validation/df_%s_mi_%i_fold_%i.rds" % (s, i, j)
-          for s in ("train", "test")
-          for i in range(1, config["mi_m"] + 1)
-          for j in range(1, config["folds"] + 1)
-        ]
-
 
 
 
@@ -125,26 +111,6 @@ rule model_posteriors:
           for j in range(1, config["folds"] + 1)
         ]
 
-rule model_assessment:
-    input:
-        pop_report = rules.prepare_data.output,
-        posteriors = rules.model_posteriors.input,
-        markdown   = "reports/model_assessment.Rmd"
-    output:
-        pdf     = "output/{version}/model_assessment.pdf",
-        figures = "output/{version}/model_assessment_figures.zip"
-    shell:
-        """
-        mkdir -p output/{wildcards.version}
-        Rscript -e "rmarkdown::render(\\"{input.markdown}\\", output_dir = \\"output/{wildcards.version}\\", params = list(data_dir = \\"../output/{wildcards.version}/data\\", config_file = \\"../config.yml\\"))"
-        mv reports/figures.zip {output.figures}
-        """
-
-# define corresponding target rule for ease of use
-rule cv_model_comparison_report_v1_1:
-    input:
-        pdf     = "output/v1.1/model_assessment.pdf",
-        figures = "output/v1.1/model_assessment_figures.zip"
 
 
 

download.R

+#!/usr/bin bash
+
+curl \
+  --user $NEUROBOT_USR:$NEUROBOT_API \
+  --digest https://neurobot-stage.incf.org/api/data/_5c8a757252dc3879e3b7cc35.csv
+

scripts/download_v1.1.sh

+#!/usr/bin bash
+
+VERSION=v1.1
+OUT=data/$VERSION
+
+mkdir -p $OUT
+
+# ctmri
+curl \
+  --user $NEUROBOT_USR:$NEUROBOT_API \
+  --digest https://center-tbi.incf.org/api/data/_5c5485306b3f2f22e14d209a.csv > \
+  $OUT/df_ctmri.csv
+Rscript -e "library(tidyverse); saveRDS(as_tibble(read_csv('$OUT/df_ctmri.csv')), file = '$OUT/df_ctmri.rds')"
+
+# imaging
+curl \
+  --user $NEUROBOT_USR:$NEUROBOT_API \
+  --digest https://center-tbi.incf.org/api/data/_5c5488246b3f2f22e14d209d.csv > \
+  $OUT/df_imaging.csv
+Rscript -e "library(tidyverse); saveRDS(as_tibble(read_csv('$OUT/df_imaging.csv')), file = '$OUT/df_imaging.rds')"
+
+# labs
+curl \
+  --user $NEUROBOT_USR:$NEUROBOT_API \
+  --digest https://center-tbi.incf.org/api/data/_5c5489696b3f2f22e14d209f.csv > \
+  $OUT/df_labs.csv
+Rscript -e "library(tidyverse); saveRDS(as_tibble(read_csv('$OUT/df_labs.csv')), file = '$OUT/df_labs.rds')"
+
+# GOSe
+curl \
+  --user $NEUROBOT_USR:$NEUROBOT_API \
+  --digest https://center-tbi.incf.org/api/data/_5c548a056b3f2f22e14d20a0.csv > \
+  $OUT/df_gose.csv
+Rscript -e "library(tidyverse); saveRDS(as_tibble(read_csv('$OUT/df_gose.csv')), file = '$OUT/df_gose.rds')"
+
+# baseline
+curl \
+  --user $NEUROBOT_USR:$NEUROBOT_API \
+  --digest https://center-tbi.incf.org/api/data/_5c548a5b6b3f2f22e14d20a2.csv > \
+  $OUT/df_baseline.csv
+Rscript -e "library(tidyverse); saveRDS(as_tibble(read_csv('$OUT/df_baseline.csv')), file = '$OUT/df_baseline.rds')"

scripts/import_neurobot_data.R

-library(tidyverse)
-
-args <- commandArgs(trailingOnly = TRUE)
-
-in_folder  <- args[[1]]
-out_folder <- args[[2]]
-
-in_files <- sprintf("%s/%s", in_folder, list.files(path = in_folder, pattern = "*.csv"))
-
-if (length(in_files) != 5) stop("must have exactly 5 input files")
-
-for (f in in_files) {
-  tmp <- read_csv(f)
-  if ("Subject.Age" %in% names(tmp))
-    saveRDS(as_tibble(tmp), file = sprintf("%s/df_baseline.rds", out_folder))
-  if ("Outcomes.DerivedCompositeGOSE" %in% names(tmp))
-    saveRDS(as_tibble(tmp), file = sprintf("%s/df_gose.rds", out_folder))
-  if ("Labs.DLDate" %in% names(tmp))
-    saveRDS(as_tibble(tmp), file = sprintf("%s/df_labs.rds", out_folder))
-  if ("Imaging.MarshallCTClassification" %in% names(tmp))
-    saveRDS(as_tibble(tmp), file = sprintf("%s/df_imaging.rds", out_folder))
-  if ("CTMRI.CTSubarachnoidHem" %in% names(tmp))
-    saveRDS(as_tibble(tmp), file = sprintf("%s/df_ctmri.rds", out_folder))
-}