Add PlasFlow and RGI

merge checkpoint_snakefile_VG into checkpoint_snakefile to include PlasFlow and RGI

Note: The config file and main snakemake file have been adjusted. Further changes should be made in other snakemake files as only the files for the analysis step have been adjusted.

See commits mentioned in the issue #28 (closed) and #29 (closed).

2019_GDB/rules/ANALYSIS_RULES

@@ -233,6 +233,7 @@ rule plasflow:
 # RGI #
 #######
 # RGI input: proteins
+# NOTE: remove stop codon symbol "*"
 # NOTE: one rule per assembly to have a workaround for the issue with file paths
 #       should be resolved properly later
 rule rgi_input_flye:
@@ -247,7 +248,7 @@ rule rgi_input_metaspades_hybrid:
     input:
         os.path.abspath(os.path.join(RESULTS_DIR, "annotation/proteins/metaspades_hybrid/lr_no_barcode-sr_ONT3_MG_xx_Rashi_S11/contigs.faa"))
     output:
-        os.path.join(RESULTS_DIR, "analysis/rgi/metaspades_hybrid.faa")
+        temp(os.path.join(RESULTS_DIR, "analysis/rgi/metaspades_hybrid.faa"))
     shell:
         "sed 's/\*$//' {input} > {output}"
 
@@ -255,7 +256,7 @@ rule rgi_input_metaspades:
     input:
         os.path.abspath(os.path.join(RESULTS_DIR, "annotation/proteins/metaspades/ONT3_MG_xx_Rashi_S11/final.contigs.faa"))
     output:
-        os.path.join(RESULTS_DIR, "analysis/rgi/metaspades.faa")
+        temp(os.path.join(RESULTS_DIR, "analysis/rgi/metaspades.faa"))
     shell:
         "sed 's/\*$//' {input} > {output}"
 
@@ -263,7 +264,7 @@ rule rgi_input_megahit:
     input:
         os.path.abspath(os.path.join(RESULTS_DIR, "annotation/proteins/megahit/ONT3_MG_xx_Rashi_S11/final.contigs.faa"))
     output:
-        os.path.join(RESULTS_DIR, "analysis/rgi/megahit.faa")
+        temp(os.path.join(RESULTS_DIR, "analysis/rgi/megahit.faa"))
     shell:
         "sed 's/\*$//' {input} > {output}"
 
@@ -300,10 +301,11 @@ rule rgi_prot:
     message:
         "AMR prediction w/ RGI: {input}"
     shell:
+        # NOTE: to make sure that the correct DB is used
         "rgi clean --local &> {log} && "
         "rgi load --card_json {input.db} --local &>> {log} && "
         "rgi database --version --local &>> {log} && "
-        # NOTE: https://github.com/arpcard/rgi/issues/93
+        # NOTE: https://github.com/arpcard/rgi/issues/93: KeyError: 'snp'
         #       need to run the CMD twice
         "rgi main --input_sequence {input.faa} --output_file {params.obname} --input_type protein --local -a {params.alignment_tool} --clean -n {threads} &>> {log} || "
         "rgi main --input_sequence {input.faa} --output_file {params.obname} --input_type protein --local -a {params.alignment_tool} --clean -n {threads} &>> {log}"