[GEN-1574] Add merge and decode to staging

.github/workflows/build-docker-images.yml

@@ -13,7 +13,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        module: ["references", "table_updates"] # Define the modules you want to loop through for builds
+        module: ["references", "table_updates", "uploads"] # Define the modules you want to loop through for builds
     env:
       REGISTRY: ghcr.io
       IMAGE_NAME: sage-bionetworks/genie-bpc-pipeline

main.nf

@@ -93,14 +93,17 @@ workflow BPC_PIPELINE {
    ch_comment = Channel.value(params.comment)
 
    if (params.step == "update_potential_phi_fields_table") {
-    update_potential_phi_fields_table(ch_comment, params.production)// validate_data.out.view()
-    } else if (params.step == "update_data_table") {
+    update_potential_phi_fields_table(ch_comment, params.production)
+    // validate_data.out.view()
+   } else if (params.step == "merge_and_uncode_rca_uploads"){
+    merge_and_uncode_rca_uploads("default", ch_cohort, ch_comment, params.production)
+   } else if (params.step == "update_data_table") {
     update_data_table("default", ch_cohort, ch_comment, params.production)
-    } else if (params.step == "genie_bpc_pipeline"){
+   } else if (params.step == "genie_bpc_pipeline"){
     update_potential_phi_fields_table(ch_comment, params.production)
     run_quac_upload_report_error(update_potential_phi_fields_table.out, ch_cohort)
     run_quac_upload_report_warning(run_quac_upload_report_error.out, ch_cohort, params.production)
-    merge_and_uncode_rca_uploads(run_quac_upload_report_warning.out, ch_cohort, params.production)
+    merge_and_uncode_rca_uploads(run_quac_upload_report_warning.out, ch_cohort, ch_comment, params.production)
     // remove_patients_from_merged(merge_and_uncode_rca_uploads.out, ch_cohort, params.production)
     update_data_table(merge_and_uncode_rca_uploads.out, ch_cohort, ch_comment, params.production)
     update_date_tracking_table(update_data_table.out, ch_cohort, ch_comment, params.production)

modules/merge_and_uncode_rca_uploads.nf

@@ -3,13 +3,14 @@ Merge and uncode REDcap export data files.
 */
 process merge_and_uncode_rca_uploads {
 
-   container 'sagebionetworks/genie-bpc-pipeline-uploads'
+   container "$params.uploads_docker"
    secret 'SYNAPSE_AUTH_TOKEN'
    debug true
 
    input:
    val previous
    val cohort
+   val comment
    val production
 
    output:
@@ -19,13 +20,13 @@ process merge_and_uncode_rca_uploads {
    if (production) {
       """
       cd /usr/local/src/myscripts/
-      Rscript merge_and_uncode_rca_uploads.R -c $cohort -u -v
+      Rscript merge_and_uncode_rca_uploads.R -c $cohort -v --production --save_synapse --comment $comment
       """
    }
    else {
       """
       cd /usr/local/src/myscripts/
-      Rscript merge_and_uncode_rca_uploads.R -c $cohort -v
+      Rscript merge_and_uncode_rca_uploads.R -c $cohort -v --save_synapse --comment $comment
       """
    }
 }

nextflow.config

@@ -53,6 +53,7 @@ profiles {
 		params {
 			// docker image parameters, see nextflow_schema.json for details
 			references_docker = "sagebionetworks/genie-bpc-pipeline-references"
+			uploads_docker = "sagebionetworks/genie-bpc-pipeline-uploads"
 			table_updates_docker = "sagebionetworks/genie-bpc-pipeline-table-updates"
 		}
 	}

nextflow_schema.json

@@ -50,14 +50,19 @@
                     "description": "Available BPC steps",
                     "enum": [
                         "update_potential_phi_fields_table",
-                        "genie_bpc_pipeline",
-                        "update_data_table"
+                        "merge_and_uncode_rca_uploads",
+                        "update_data_table",
+                        "genie_bpc_pipeline"
                     ]
                 },
                 "references_docker":{
                     "type": "string",
                     "description": "Name of docker to use in processes in scripts/references"
                 },
+                "uploads_docker":{
+                    "type": "string",
+                    "description": "Name of docker to use in processes in scripts/uploads"
+                },
                 "table_updates_docker":{
                     "type": "string",
                     "description": "Name of docker to use in processes in scripts/table_updates" 

scripts/uploads/Dockerfile

@@ -1,21 +1,34 @@
-FROM r-base:4.0.0
+FROM rstudio/r-base:4.0-bullseye
 
+# Set working directory
 WORKDIR /usr/local/src/myscripts
 
+# Set environment variable for renv version
 ENV RENV_VERSION 0.14.0
 
-RUN rm /etc/apt/apt.conf.d/default
-RUN apt-get update -y
-RUN apt-get install -y dpkg-dev zlib1g-dev libssl-dev libffi-dev
-# procps is required for nextflow tower
-RUN apt-get install -y curl libcurl4-openssl-dev procps
-RUN R -e "install.packages('synapser', repos=c('http://ran.synapse.org', 'http://cran.fhcrc.org'))"
+# Update apt-get and install system dependencies (only install required)
+RUN apt-get update -y && \
+    apt-get install -y --no-install-recommends \ 
+    dpkg-dev zlib1g-dev libssl-dev libffi-dev \
+    libcurl4-openssl-dev curl procps && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
 
-ENV PYTHON /usr/local/lib/R/site-library/PythonEmbedInR/bin/python3.6
+# Install R packages including remotes and renv
+RUN R -e "install.packages('remotes', repos = 'https://cloud.r-project.org')" && \
+    R -e "remotes::install_github('rstudio/renv', ref = '${RENV_VERSION}')" || true
 
-RUN R -e "install.packages('remotes', repos = c(CRAN = 'https://cloud.r-project.org'))"
-RUN R -e "remotes::install_github('rstudio/renv@${RENV_VERSION}')"
+# Install synapser with specific version
+RUN R -e "remotes::install_version('synapser', version = '0.11.7', repos = c('http://ran.synapse.org', 'http://cran.fhcrc.org'))"
 
-COPY . .
+# Set Python environment variable for R
+ENV PYTHON /usr/local/lib/R/site-library/PythonEmbedInR/bin/python3.6
 
+# Copy only renv.lock first to leverage docker cache for dependencies
+COPY renv.lock renv.lock
+
+# Restore R environment with renv
 RUN R -e "renv::restore()"
+
+# Copy the local project files into the container
+COPY . .

scripts/uploads/README.md

@@ -2,6 +2,10 @@
 [![automated](https://img.shields.io/docker/cloud/automated/sagebionetworks/genie-bpc-pipeline-uploads)](https://hub.docker.com/r/sagebionetworks/genie-bpc-pipeline-uploads)
 ![status](https://img.shields.io/docker/cloud/build/sagebionetworks/genie-bpc-pipeline-uploads)
 
+## Setup
+
+You will want to run on an EC2 instance with docker installed.
+
 ## Installation
 
 Clone this repository and navigate to the directory:
@@ -10,11 +14,43 @@ git clone [email protected]:Sage-Bionetworks/genie-bpc-pipeline.git
 cd genie-bpc-pipeline/bpc/uploads/
 ```
 
-Install all required R packages:
+### Local installation
+
+Install all required R packages by copying and running the commands in the Dockerfile
+at `scripts/uploads/Dockerfile` except for the last step which is copying your current repo.
+
+### Docker
+
+Alternatively, you can pull the docker image associated with this module from [here](https://github.com/Sage-Bionetworks/genie-bpc-pipeline/pkgs/container/genie-bpc-pipeline) into your EC2.
+
+You can also pull the production docker:
+
+```bash
+git pull sagebionetworks/genie-bpc-pipeline-uploads
 ```
-R -e 'renv::restore()'
+
+To run the docker image:
+
+1. Run the docker
+
+```bash
+docker run -d --name <nickname_for_container> <container_name> /bin/bash -c "while true; do sleep 1; done"
+```
+
+2. **Optional.** Do anything you need to do to the container (e.g: copy current local changes to the docker)
+
+```bash
+docker cp ./. test_container:/usr/local/src/myscripts
+```
+
+3. Execute container into a bash session
+
+```bash
+docker exec -it <nickname_for_container> /bin/bash
 ```
 
+Now you can run the script commands
+
 ## Synapse credentials
 
 Cache your Synapse personal access token (PAT) as an environmental variable:
@@ -45,7 +81,7 @@ Usage: merge_and_uncode_rca_uploads.R [options]
 Options:
         -c COHORT, --cohort=COHORT
                 BPC cohort
-
+        
         -u, --save_synapse
                 Save output to Synapse
 
@@ -54,13 +90,21 @@ Options:
 
         -h, --help
                 Show this help message and exit
+
+        --production
+                Whether to run in production mode or not (staging mode).
 ```
 
-Example run: 
+Example run (staging run):
 ```
 Rscript merge_and_uncode_rca_uploads.R -c NSCLC -u -a $SYNAPSE_AUTH_TOKEN
 ```
 
+Example run (production run):
+```
+Rscript merge_and_uncode_rca_uploads.R -c NSCLC -u -a $SYNAPSE_AUTH_TOKEN --production
+```
+
 ## Usage: remove patient IDs from a REDCap formatted file
 
 To display the command line interface:
@@ -100,3 +144,19 @@ Example run:
 ```
 Rscript remove_patients_from_merged.R -i syn23285494 -c NSCLC -r syn29266682 -v
 ```
+
+## Running tests
+There are unit tests under `scripts/uploads/tests`.
+
+1. Install the `mockery` and `testthat` packages via the command line:
+
+```bash
+R -e "remotes::install_cran('testthat')"
+```
+
+2. Run the following in a R session in your EC2:
+
+```R
+library(testthat)
+test_dir("/usr/local/src/myscripts/tests")
+```

scripts/uploads/config.yaml

@@ -10,7 +10,8 @@ synapse:
     type: table
     description: PRISSMM formatted data dictionaries for REDCap
   rca_files: 
-    id: syn23286928
+    production_id: syn23286928
+    staging_id: syn63887337
     name: curated
     type: folder
     description: REDCap academic formatted files of merged BPC data ready for import to Synapse tables