Fixed all shellcheck issues

.vscode/settings.json

@@ -1,5 +1,7 @@
 {
     "cSpell.words": [
-        "cowusda"
+        "BIOMV",
+        "cowusda",
+        "numcols"
     ]
 }

main.sh

@@ -23,14 +23,14 @@ module load r
 # Metaxa2 uses 4 cpus, we need to make sure that each instance has
 # enough cpus to run
 echo "--^-- X: Reading FASTQ sequences..."
-find . -maxdepth 1 -name "*R1_001.fastq.gz" | \
+find . -maxdepth 1 -0 -name "*R1_001.fastq.gz" | \
   xargs -L1 -P"$SLURM_NTASKS" srun -n1 -N1 --exclusive ./fastq-to-taxonomy.sh
 echo "--^-- X: Reading FASTQ sequences...Done!"
 
 # Compile those pesky individual taxonomic tables into a single
 # OTU feature table
 echo "--^-- X: Compiling feature table..."
-metaxa2_dc -i *.level_7.txt -o metaxa-feature-table.tsv
+metaxa2_dc -i ./*.level_7.txt -o metaxa-feature-table.tsv
 echo "--^-- X: Compiling feature table...Done!"
 
 # Rearrange the feature table to something QIIME likes a little bit better
@@ -49,29 +49,33 @@ MINRAREFACTION=$(<rarefaction.min.txt)
 MAXRAREFACTION=$(<rarefaction.max.txt)
 
 # Start up QIIME
+# This code creates errors if run through shellcheck because
+# the shellcheck program doesn't understand miniconda:
+# We'll add a directive to tell it to ignore this error
+# shellcheck disable=SC1091
 source activate qiime2
 
 # Convert the feature table into BIOM format
 echo "--^-- X: Importing data..."
 biom convert \
--i feature-table.tsv \
--o feature-table.hdf5.biom \
---table-type="OTU table" \
---to-hdf5 \
---process-obs-metadata taxonomy
+  -i feature-table.tsv \
+  -o feature-table.hdf5.biom \
+  --table-type="OTU table" \
+  --to-hdf5 \
+  --process-obs-metadata taxonomy
 
 # Now convert the BIOM table into QIIME format (good grief!)
 qiime tools import \
- --input-path feature-table.hdf5.biom \
- --type 'FeatureTable[Frequency]' \
- --input-format 'BIOMV210Format' \
- --output-path feature-table.qza
- 
+  --input-path feature-table.hdf5.biom \
+  --type 'FeatureTable[Frequency]' \
+  --input-format 'BIOMV210Format' \
+  --output-path feature-table.qza
+
 qiime tools import \
- --input-path feature-table.hdf5.biom \
- --output-path taxonomy.qza \
- --type 'FeatureData[Taxonomy]' \
- --input-format 'BIOMV210Format' 
+  --input-path feature-table.hdf5.biom \
+  --output-path taxonomy.qza \
+  --type 'FeatureData[Taxonomy]' \
+  --input-format 'BIOMV210Format'
 echo "--^-- X: Importing data...Done!"
 
 # We will need to run core-metrics to generate information further down
@@ -79,12 +83,12 @@ echo "--^-- X: Running core-metrics..."
 rm -r "core-metrics-results"
 # This is one of the few QIIME commands that can use multithreading
 qiime diversity core-metrics \
- --i-table feature-table.qza \
- --p-sampling-depth "$MINRAREFACTION" \
- --m-metadata-file metadata.tsv \
- --p-n-jobs 4 \
- --output-dir core-metrics-results \
- --verbose
+  --i-table feature-table.qza \
+  --p-sampling-depth "$MINRAREFACTION" \
+  --m-metadata-file metadata.tsv \
+  --p-n-jobs 4 \
+  --output-dir core-metrics-results \
+  --verbose
 echo "--^-- X: Running core-metrics...Done!"
 
 # Clean out the visualizations, or else QIIME will throw a fit
@@ -94,81 +98,80 @@ mkdir visualizations
 # Create a pretty barplot as a reward for all that effort
 echo "--^-- X: Generating barplot..."
 qiime taxa barplot \
- --i-table feature-table.qza \
- --i-taxonomy taxonomy.qza \
- --m-metadata-file metadata.tsv \
- --o-visualization visualizations/barplot.qzv
+  --i-table feature-table.qza \
+  --i-taxonomy taxonomy.qza \
+  --m-metadata-file metadata.tsv \
+  --o-visualization visualizations/barplot.qzv
 echo "--^-- X: Generating barplot...Done!"
 
 echo "--^-- X: Plotting rarefaction curve..."
 # Create a rarefaction curve to make sure the magic of rarefaction is valid
 qiime diversity alpha-rarefaction \
- --i-table feature-table.qza \
- --p-max-depth "$MAXRAREFACTION" \
- --m-metadata-file metadata.tsv \
- --o-visualization visualizations/rarefaction-curve.qzv
+  --i-table feature-table.qza \
+  --p-max-depth "$MAXRAREFACTION" \
+  --m-metadata-file metadata.tsv \
+  --o-visualization visualizations/rarefaction-curve.qzv
 echo "--^-- X: Plotting rarefaction curve...Done!"
 
 # Run alpha-diversity group significance: this will automatically include all the columns
 # Evenness first
 echo "--^-- X: Finding alpha-group-significance..."
 qiime diversity alpha-group-significance \
- --i-alpha-diversity core-metrics-results/evenness_vector.qza \
- --m-metadata-file metadata.tsv \
- --o-visualization visualizations/evenness-group-significance.qzv \
- --verbose
-  
+  --i-alpha-diversity core-metrics-results/evenness_vector.qza \
+  --m-metadata-file metadata.tsv \
+  --o-visualization visualizations/evenness-group-significance.qzv \
+  --verbose
+
 # Now richness
 qiime diversity alpha-group-significance \
- --i-alpha-diversity core-metrics-results/shannon_vector.qza \
- --m-metadata-file metadata.tsv \
- --o-visualization visualizations/shannon-group-significance.qzv \
- --verbose
+  --i-alpha-diversity core-metrics-results/shannon_vector.qza \
+  --m-metadata-file metadata.tsv \
+  --o-visualization visualizations/shannon-group-significance.qzv \
+  --verbose
 echo "--^-- X: Finding alpha-group-significance...Done!"
-  
+
 # Now let's find the correlation between alpha-diversity and the numeric traits
 echo "--^-- X: Finding alpha-correlations..."
 qiime diversity alpha-correlation \
- --i-alpha-diversity core-metrics-results/evenness_vector.qza \
- --m-metadata-file metadata.tsv \
- --p-method pearson \
- --o-visualization visualizations/evenness-correlation.qzv \
- --verbose
- 
+  --i-alpha-diversity core-metrics-results/evenness_vector.qza \
+  --m-metadata-file metadata.tsv \
+  --p-method pearson \
+  --o-visualization visualizations/evenness-correlation.qzv \
+  --verbose
+
 qiime diversity alpha-correlation \
- --i-alpha-diversity core-metrics-results/shannon_vector.qza \
- --m-metadata-file metadata.tsv \
- --p-method pearson \
- --o-visualization visualizations/shannon-correlation.qzv \
- --verbose
+  --i-alpha-diversity core-metrics-results/shannon_vector.qza \
+  --m-metadata-file metadata.tsv \
+  --p-method pearson \
+  --o-visualization visualizations/shannon-correlation.qzv \
+  --verbose
 echo "--^-- X: Finding alpha-correlations...Done!"
 
 # Now for the tricky part: beta-diversity
 echo "--^-- X: Checking entries for beta-significance..."
 # QIIME only uses one processor for these, so we can parallelize this step
-cat catcols.txt | \
-  xargs -P"$SLURM_NTASKS" -I {} srun -n1 -N1 --exclusive \
+xargs -a catcols.txt -P"$SLURM_NTASKS" -I {} srun -n1 -N1 --exclusive \
   qiime diversity beta-group-significance \
     --i-distance-matrix core-metrics-results/bray_curtis_distance_matrix.qza \
-	--m-metadata-file metadata.tsv \
-	--m-metadata-column {} \
-	--p-pairwise \
-	--o-visualization "visualizations/bray-curtis-{}-significance.qzv" \
-	--verbose
+    --m-metadata-file metadata.tsv \
+    --m-metadata-column {} \
+    --p-pairwise \
+    --o-visualization "visualizations/bray-curtis-{}-significance.qzv" \
+    --verbose
+
 echo "--^-- X: Checking entries for beta-significance...Done!"
 
 echo "--^-- X: Performing ANCOM..."
 # We will try to use ancom on the full dataset, although it might kill us
 # Extract pseudocount
 qiime composition add-pseudocount \
- --i-table feature-table.qza \
- --o-composition-table composition-table.qza
+  --i-table feature-table.qza \
+  --o-composition-table composition-table.qza
 
 # Run ancom for all categories in catcols
 # Once again, QIIME only uses one processor (even though this
 # is a HUGE task), so we should parallelize it for speed
-cat catcols.txt | \
-  xargs -P"$SLURM_NTASKS" -I {} srun -n1 -N1 --exclusive \
+xargs -a catcols.txt -P"$SLURM_NTASKS" -I {} srun -n1 -N1 --exclusive \
   qiime composition ancom \
    --i-table composition-table.qza \
    --m-metadata-file metadata.tsv \
@@ -178,13 +181,11 @@ cat catcols.txt | \
 echo "--^-- X: Performing ANCOM...Done!"
 
 # Create category-based predictive models
-cat catcols.txt | \
-  xargs -P"$SLURM_NTASKS" -L1 srun -n1 -N1 --exclusive \
+xargs -a catcols.txt -P"$SLURM_NTASKS" -L1 srun -n1 -N1 --exclusive \
   ./sample-classifier.sh
 
 # Create continuous predictive models
-cat numcols.txt | \
-  xargs -P"$SLURM_NTASKS" -L1 srun -n1 -N1 --exclusive \
+xargs -a numcols.txt -P"$SLURM_NTASKS" -L1 srun -n1 -N1 --exclusive \
   ./sample-regression.sh
 
-echo "All Done!"
+echo "--^-- X: All Done!"

sample-classifier.sh

@@ -16,9 +16,14 @@ module load swset
 module load miniconda3
 
 # Start up qiime
+# This code creates errors if run through shellcheck because
+# the shellcheck program doesn't understand miniconda:
+# We'll add a directive to tell it to ignore this error
+# shellcheck disable=SC1091
 source activate qiime2
 
 # Make sure we have a clean slate to work with
+echo "--^-- X: Creating a classifier model for ${CATEGORY}"
 echo "--^-- X: Clearing previous classifier results..."
 rm -r "$FOLDERNAME"
 echo "--^-- X: Clearing previous classifier results...Done!"
@@ -26,24 +31,24 @@ echo "--^-- X: Clearing previous classifier results...Done!"
 # Solve the model
 echo "--^-- X: Constructing model..."
 qiime sample-classifier classify-samples \
- --i-table feature-table.qza \
- --m-metadata-file metadata.tsv \
- --m-metadata-column "$CATEGORY" \
- --p-n-jobs 4 \
- --p-missing-samples ignore \
- --p-optimize-feature-selection \
- --output-dir "$FOLDERNAME" \
- --verbose
+  --i-table feature-table.qza \
+  --m-metadata-file metadata.tsv \
+  --m-metadata-column "$CATEGORY" \
+  --p-n-jobs 4 \
+  --p-missing-samples ignore \
+  --p-optimize-feature-selection \
+  --output-dir "$FOLDERNAME" \
+  --verbose
 echo "--^-- X: Constructing model...Done!"
 
 # Convert the model output into readable visualizations
 echo "--^-- X: Making visualizations..."
 qiime metadata tabulate \
- --m-input-file "${FOLDERNAME}/feature_importance.qza" \
- --o-visualization "${FOLDERNAME}/feature-importance.qzv"
+  --m-input-file "${FOLDERNAME}/feature_importance.qza" \
+  --o-visualization "${FOLDERNAME}/feature-importance.qzv"
 
 qiime metadata tabulate \
- --m-input-file "${FOLDERNAME}/predictions.qza" \
- --m-input-file metadata.tsv \
- --o-visualization "${FOLDERNAME}/predictions.qzv"
+  --m-input-file "${FOLDERNAME}/predictions.qza" \
+  --m-input-file metadata.tsv \
+  --o-visualization "${FOLDERNAME}/predictions.qzv"
 echo "--^-- X: Making visualizations...Done!"

sample-regression.sh

@@ -16,9 +16,14 @@ module load swset
 module load miniconda3
 
 # Start up qiime
+# This code creates errors if run through shellcheck because
+# the shellcheck program doesn't understand miniconda:
+# We'll add a directive to tell it to ignore this error
+# shellcheck disable=SC1091
 source activate qiime2
 
 # Make sure we have a clean slate to work with
+echo "--^-- X: Creating regression model for ${CATEGORY}"
 echo "--^-- X: Clearing previous regression results..."
 rm -r "$FOLDERNAME"
 echo "--^-- X: Clearing previous regression results...Done!"
@@ -26,24 +31,24 @@ echo "--^-- X: Clearing previous regression results...Done!"
 # Solve the model
 echo "--^-- X: Constructing model..."
 qiime sample-classifier regress-samples \
- --i-table feature-table.qza \
- --m-metadata-file metadata.tsv \
- --m-metadata-column "$CATEGORY" \
- --p-n-jobs 4 \
- --p-missing-samples ignore \
- --p-optimize-feature-selection \
- --output-dir "$FOLDERNAME" \
- --verbose
+  --i-table feature-table.qza \
+  --m-metadata-file metadata.tsv \
+  --m-metadata-column "$CATEGORY" \
+  --p-n-jobs 4 \
+  --p-missing-samples ignore \
+  --p-optimize-feature-selection \
+  --output-dir "$FOLDERNAME" \
+  --verbose
 echo "--^-- X: Constructing model...Done!"
 
 # Convert the model output into readable visualizations
 echo "--^-- X: Making visualizations..."
 qiime metadata tabulate \
- --m-input-file "${FOLDERNAME}/feature_importance.qza" \
- --o-visualization "${FOLDERNAME}/feature-importance.qzv"
+  --m-input-file "${FOLDERNAME}/feature_importance.qza" \
+  --o-visualization "${FOLDERNAME}/feature-importance.qzv"
 
 qiime metadata tabulate \
- --m-input-file "${FOLDERNAME}/predictions.qza" \
- --m-input-file metadata.tsv \
- --o-visualization "${FOLDERNAME}/predictions.qzv"
+  --m-input-file "${FOLDERNAME}/predictions.qza" \
+  --m-input-file metadata.tsv \
+  --o-visualization "${FOLDERNAME}/predictions.qzv"
 echo "--^-- X: Making visualizations...Done!"