v2.5.5

Dockerfile

@@ -2,6 +2,8 @@ FROM condaforge/mambaforge:latest
 LABEL io.github.snakemake.containerized="true"
 LABEL io.github.snakemake.conda_env_hash="3e0f6a910323a150167cddb17cdf65d15adb02b0502443384c88795d2a6fc76e"
 
+# v2.5.5
+
 # Step 1: Retrieve conda environments
 
 # Conda environment:

README.md

@@ -144,7 +144,7 @@ Finally, define the alias that will be used to launch asscom2 from any directory
 ```bash
 echo "export ASSCOM2_BASE=$(pwd -P)" >> ~/.bashrc # Save installation directory. 
 echo "export ASSCOM2_PROFILE=\${ASSCOM2_BASE}/profiles/apptainer/local" >> ~/.bashrc # Define profile selection.
-echo "export ASSCOM2_DATABASES=${ASSCOM2_BASE}/databases" >> ~/.bashrc # Define database base directory.
+echo "export ASSCOM2_DATABASES=\${ASSCOM2_BASE}/databases" >> ~/.bashrc # Define database base directory.
 
 echo "alias asscom2='conda run --live-stream --name asscom2_launcher snakemake --snakefile \${ASSCOM2_BASE}/snakefile --profile \${ASSCOM2_PROFILE} --configfile \${ASSCOM2_BASE}/config.yaml'" >> ~/.bashrc
 source ~/.bashrc

changelog.txt

@@ -1,3 +1,7 @@
+2.5.5
+  - Use $ASSCOM_DATABASES to define where the databases are placed. 
+  - Docker images have a versioning system so an incompatible image can't be used.
+
 2.5.4
   - Using the mashtree order in plots in the report. Using alphabetical order redundantly in case mashtree hasn't run (sucessfully). Minor adjustments to other plots. TODO: midpoint root tree somehow and order all tables in the report as well.
 

profiles/apptainer/local/config.yaml

@@ -12,8 +12,9 @@ keep-incomplete: true # true only for debugging
 # Apptainer
 use-conda: true # Omitting this line was the culprit, now the 'snakemake --containerize' auto stuff works!
 use-singularity: true
-# What happens if I don't define the singularity-prefix? Does it download the image to the running directory of the alias?
-#singularity-prefix: '~/.asscom2/apptainer-prefix' # Directory in which singularity images will be stored. Couldn't find a way of using the $ASSCOM_BASE system variable.
+
+# If you don't define a common singularity-prefix, the image will be pulled every single time
+singularity-prefix: "~/.asscom2/singularity-prefix" # Directory in which singularity images will be stored. Couldn't find a way of using the $ASSCOM_BASE system variable, so users should manually change this here if they want to use something else.
 singularity-args: "--bind $ASSCOM2_BASE,$ASSCOM2_DATABASES" # Must also bind ASSCOM2_BASE so we can access databases and report scripts etc. When this issue is solved I'll add individual arguments to rules and make it more flexible https://github.com/snakemake/snakemake/issues/262
 
 # Comments for the singularity-args parameter. I don't know why the necessary /opt/conda/bin path, which is present when I run the image, is gone when snakemake runs. Also, since the databases are likely in a soft linked directory, you must bind it if you want access

report_subpipeline/scripts/genomes_to_report_v3.Rmd

@@ -434,5 +434,5 @@ section_definition_table %>%
 
 
 
-<p class = "shaded">[assemblycomparator2 v2.5.4](https://github.com/cmkobel/assemblycomparator2#assemblycomparator2) genomes to report pipeline. Copyright (C) 2019-2023 [Carl M. Kobel](https://github.com/cmkobel) [GNU GPL v3](https://github.com/cmkobel/assemblycomparator2/blob/master/LICENSE)</p>
+<p class = "shaded">[assemblycomparator2 v2.5.5](https://github.com/cmkobel/assemblycomparator2#assemblycomparator2) genomes to report pipeline. Copyright (C) 2019-2023 [Carl M. Kobel](https://github.com/cmkobel) [GNU GPL v3](https://github.com/cmkobel/assemblycomparator2/blob/master/LICENSE)</p>
 

report_subpipeline/snakefile

@@ -2,7 +2,7 @@
 
 containerized: "docker://cmkobel/assemblycomparator2"
 
-__version__ = "v2.5.4" 
+__version__ = "v2.5.5" 
 __author__ = 'Carl M. Kobel'
 
 

snakefile

@@ -4,7 +4,7 @@
 # asscom2 --containerize > Dockerfile # And remove header-text.
 
 
-__version__ = "v2.5.4" # Four places to bump. Here, in the bottom of the report, in the report snakefile. And in the changelog.
+__version__ = "v2.5.5" # Five places to bump. Here, in the bottom of the report, in the report snakefile, in the changelog and the Dockerfile.
 __author__ = 'Carl M. Kobel'
 
 # May the data passing through this pipeline
@@ -128,6 +128,12 @@ print()
 
 
 
+# The DATABASES directory must exist, otherwise apptainer gets confused and throws:
+# WARNING: skipping mount of /home/thylakoid/assemblycomparator2/adatabaseas: stat /home/thylakoid/assemblycomparator2/adatabaseas: no such file or directory
+if not os.path.isdir(DATABASES):
+    os.mkdir(DATABASES)
+
+
 # --- Make sure the output directory exists. ----------------------------------
 
 """ try: 
@@ -252,7 +258,7 @@ rule busco_download:
             # https://busco.ezlab.org/busco_userguide.html#download-and-automated-update
             busco \
                 --in dummy.fasta \
-                --out shouldnotbenecessarytosettheoutdirwhenjustdownloading \
+                --out dummy \
                 --mode geno \
                 --auto-lineage-prok \
                 --force \
@@ -707,7 +713,8 @@ rule dbcan: # I can't decide whether this rule should really be called "run_dbca
     input: 
         metadata = "{results_directory}/metadata.tsv",
         aminoacid = "{results_directory}/samples/{sample}/prokka/{sample}.faa", # From prokka
-        database_representative = expand("{base_variable}/databases/dbcan/ac2_dbcan_database_representative.flag", base_variable = base_variable),
+        #database_representative = expand("{base_variable}/databases/dbcan/ac2_dbcan_database_representative.flag", base_variable = base_variable),
+        database_representative = DATABASES + "/dbcan/ac2_dbcan_database_representative.flag"
     output: 
         overview_table = "{results_directory}/samples/{sample}/dbcan/overview.txt"
     params: 
@@ -935,7 +942,9 @@ rule diamond_kegg: # or uniref_ko?
     input: 
         metadata = "{results_directory}/metadata.tsv", # For the report
         aminoacid = "{results_directory}/samples/{sample}/prokka/{sample}.faa", # From prokka
-        database_representative = base_variable + "/databases/checkm2/ac2_checkm2_database_representative.flag",
+        #database_representative = base_variable + "/databases/checkm2/ac2_checkm2_database_representative.flag",
+        database_representative = DATABASES + "/checkm2/ac2_checkm2_database_representative.flag",
+
     output:
         tsv = "{results_directory}/samples/{sample}/diamond_kegg/{sample}_diamond_kegg.tsv", # Or should it be named uniref-100?
     params: