Nextflow work

.gitlab-ci.yml

@@ -2,8 +2,8 @@
 variables:
   S3_ENDPOINT: "https://s3.embl.de"
   S3_BUCKET: "annelids"
-  JBROWSE_HOSTNAME: "genomes-dev.arendt.embl.de"
-  RELEASE_LABEL: "dev"
+  JBROWSE_HOSTNAME: "genomes-testing.arendt.embl.de"
+  RELEASE_LABEL: "testing"
   INGRESS_CLASS: "internal-users"
 
 stages:
@@ -66,6 +66,11 @@ kustomize_build_apply:
         JBROWSE_HOSTNAME: "genomes.arendt.embl.de"
         RELEASE_LABEL: "prod"
         INGRESS_CLASS: "external-users"
+    - if: '$CI_COMMIT_REF_NAME  == "dev"'
+      variables:
+        JBROWSE_HOSTNAME: "genomes-dev.arendt.embl.de"
+        RELEASE_LABEL: "dev"
+        INGRESS_CLASS: "internal-users"
     - when: always
 
 clean_up_k8s_deployment:
@@ -87,4 +92,10 @@ clean_up_k8s_deployment:
         JBROWSE_HOSTNAME: "genomes.arendt.embl.de"
         RELEASE_LABEL: "prod"
         INGRESS_CLASS: "external-users"
-    - when: on_failure
+    - if: '$CI_COMMIT_REF_NAME  == "dev"'
+      when: on_failure
+      variables:
+        JBROWSE_HOSTNAME: "genomes-dev.arendt.embl.de"
+        RELEASE_LABEL: "dev"
+        INGRESS_CLASS: "internal-users"
+    - when: always

resources_to_s3_NF/genomes.csv

-name,path
-pdumv021,/g/arendt/data/genome_assemblies/pdum-v2.1/genome-assemblies/FINAL_v021_ASM/pdumv021_genome.fa
-pmassv1_masked,/g/arendt/data/platynereis_massiliensis/genome/final_assembly/pmassv1.genome.masked.fa
\ No newline at end of file
+name,path,display_name
+pdumv021,/g/arendt/data/genome_assemblies/pdum-v2.1/genome-assemblies/FINAL_v021_ASM/pdumv021_genome.fa,platychou
+pmassv1_masked,/g/arendt/data/platynereis_massiliensis/genome/final_assembly/pmassv1.genome.masked.fa,mid platy

resources_to_s3_NF/jbrowse_script_slug.sh

 #!/bin/bash
-## Gameplan is to parse the genomes and csv files
-tail -n +1 tracks.csv | 
-sort -t , -k 1,2 | 
-awk -F, '$4=="VCF" { print "jbrowse add-track https://s3.embl.de/annelids/"$1"/variation/"$3".vcf.gz -a \""$1"\" -d \""$6"\" -n \""$3"\" --category \""$5"\""}' >> ../image/jbrowse_script.sh
-# Using this to filter above $4=="VCF"
\ No newline at end of file
+#
+# I am going to assume the CSV files have been patched with the s3 basename path - field 7 for tracks
+#
+
+# Genomes
+tail -n +1 genomes.csv | # it should not be genomes
+sort -t , -k1,1 |
+awk -F, '{ print "jbrowse add-assembly https://s3.embl.de/annelids/"$1"/"$2" \
+-n \""$1"\" --displayName \""$3"\"" }'
+
+tail -n +1 tracks.csv | # it should not be tracks
+sort -t , -k1,1 -k5,5 -k2,2 |
+awk -F, '{ print "jbrowse add-track https://s3.embl.de/annelids/"$1"/"$5"/"$3".vcf.gz \
+-a \""$1"\" -d \""$6"\" -n \""$3"\" --category \""$5"\""}'
+
+# Text-indices
+tail -n +1 tracks.csv |
+awk '$1 {print;}'