testing picrust functionality

Author: kbpi314

mmeds/snakemake/rules/file_manipulation.smk

@@ -14,6 +14,55 @@ rule extract_feature_table_tsv:
         "{output} "
         "tables/tmp_unzip_{wildcards.table}"
 
+rule extract_feature_table_biom:
+    """ Remove feature table biom file from qza archive """
+    input: 
+        "tables/{table}.qza"
+    output:
+        "tables/{table}.biom"
+    wildcard_constraints:
+        table = "[^/]+"
+    conda:
+        "mmeds_test"
+    shell:
+        "extract_feature_table.sh "
+        "{input} "
+        "{output} "
+        "biom "
+        "tables/tmp_unzip_{wildcards.table}"
+
+rule extract_feature_table_fasta:
+    """ Remove underlying fasta data from qza archive """
+    input: 
+        "tables/{table}.qza"
+    output:
+        "tables/{table}.fasta"
+    wildcard_constraints:
+        table = "[^/]+"
+    conda:
+        "mmeds_test"
+    shell:
+        "extract_feature_table.sh "
+        "{input} "
+        "{output} "
+        "fasta "
+        "tables/tmp_unzip_{wildcards.table}"
+
+rule extract_feature_table_tsv:
+    """ Remove feature table biom file from qza archive, convert to readable tsv for downstream analysis """
+    input:
+        "tables/{table}.qza"
+    output:
+        "tables/{table}.tsv"
+    conda:
+        "mmeds_test"
+    shell:
+        "extract_feature_table.sh "
+        "{input} "
+        "{output} "
+        "tsv "
+        "tables/tmp_unzip_{wildcards.table}"
+
 rule extract_feature_table_tsv_class:
     input:
         "tables/{class}/{table}.qza"

mmeds/snakemake/rules/picrust2.smk

@@ -0,0 +1,12 @@
+rule picrust2_core:
+    """ Generate picrust2 tables for input feature table """
+    input:
+        rep_seqs = "tables/rep_seqs_table.fasta"
+        biom_feature_table = "tables/asv_table.biom"
+    output:
+        picrust2_results = "picrust2_out"
+    conda:
+        # "qiime2-2020.8.0" # or should I use "qiime2-2023.9"
+        /sc/arion/projects/MMEDS/.modules/picrust2
+    shell:
+        picrust2_pipeline.py -s {input.rep_seqs} -s {input.biom_feature_table} {output.picrust2_results}

mmeds/snakemake/workflows/picrust2.Snakefile

@@ -0,0 +1,10 @@
+configfile: "config_file.yaml"
+report: "report.rst"
+
+include: "{snakemake_dir}/common.smk"
+include: "{snakemake_dir}/file_manipulation.smk"
+include: "{snakemake_dir}/table_filtering.smk"
+
+rule results:
+    input:
+        picrust2_core

scripts/extract_feature_table.sh

@@ -4,8 +4,21 @@
 
 in_file=$1
 out_file=$2
-tmp_dir_name="${3:-tmp_unzip}"
+file_type="${3:-tsv}" # tsv, biom, or fasta
+tmp_dir_name="${4:-tmp_unzip}"
 unzip -joq $in_file -d $tmp_dir_name
-biom convert --to-tsv -i $tmp_dir_name/feature-table.biom -o $out_file
-sed -i '1d;2s/^#//' $out_file
-rm -rf $tmp_dir_name
+
+if [file_type = tsv]; then
+    biom convert --to-tsv -i $tmp_dir_name/feature-table.biom -o $out_file
+    sed -i '1d;2s/^#//' $out_file
+    rm -rf $tmp_dir_name
+elif [file_type = biom]; then
+    cp $tmp_dir_name/feature-table.biom $out_file
+    rm -rf $tmp_dir_name
+elif [file_type = fasta]; then
+    cp $tmp_dir_name/dna-sequences.fasta $out_file
+    rm -rf $tmp_dir_name
+else
+    echo "Invalid file_type argument"
+fi
+