fix lints

Snakefile

@@ -41,12 +41,12 @@ rule all:
         "qc/multiqc_report.html",
 
 
-##### setup singularity #####
+##### setup container #####
 
 
-# this container defines the underlying OS for each job when using the workflow
+# this container image defines the underlying OS for each job when using the workflow
 # with --use-conda --use-singularity
-singularity: "docker://continuumio/miniconda3"
+container: "docker://continuumio/miniconda3"
 
 
 ##### setup report #####

rules/align.smk

@@ -1,18 +1,3 @@
-def get_fq(wildcards):
-    if config["trimming"]["skip"]:
-        # no trimming, use raw reads
-        return units.loc[(wildcards.sample, wildcards.unit), ["fq1", "fq2"]].dropna()
-    else:
-        # yes trimming, use trimmed data
-        if not is_single_end(**wildcards):
-            # paired-end sample
-            return expand(
-                "trimmed/{sample}-{unit}.{group}.fastq.gz", group=[1, 2], **wildcards
-            )
-        # single end sample
-        return "trimmed/{sample}-{unit}.fastq.gz".format(**wildcards)
-
-
 rule align:
     input:
         sample=get_fq,

rules/common.smk

@@ -1,2 +1,43 @@
 def is_single_end(sample, unit):
     return pd.isnull(units.loc[(sample, unit), "fq2"])
+
+
+def get_fastq(wildcards):
+    return units.loc[(wildcards.sample, wildcards.unit), ["fq1", "fq2"]].dropna()
+
+
+def get_fq(wildcards):
+    if config["trimming"]["skip"]:
+        # no trimming, use raw reads
+        return units.loc[(wildcards.sample, wildcards.unit), ["fq1", "fq2"]].dropna()
+    else:
+        # yes trimming, use trimmed data
+        if not is_single_end(**wildcards):
+            # paired-end sample
+            return expand(
+                "trimmed/{sample}-{unit}.{group}.fastq.gz", group=[1, 2], **wildcards
+            )
+        # single end sample
+        return "trimmed/{sample}-{unit}.fastq.gz".format(**wildcards)
+
+
+def get_strandedness(units):
+    if "strandedness" in units.columns:
+        return units["strandedness"].tolist()
+    else:
+        strand_list = ["none"]
+        return strand_list * units.shape[0]
+
+
+def get_deseq2_threads(wildcards=None):
+    # https://twitter.com/mikelove/status/918770188568363008
+    few_coeffs = False if wildcards is None else len(get_contrast(wildcards)) < 10
+    return 1 if len(samples) < 100 or few_coeffs else 6
+
+
+def strip_suffix(pattern, suffix):
+    return pattern[: -len(suffix)]
+
+
+def get_contrast(wildcards):
+    return config["diffexp"]["contrasts"][wildcards.contrast]

rules/diffexp.smk

@@ -1,11 +1,3 @@
-def get_strandness(units):
-    if "strandedness" in units.columns:
-        return units["strandedness"].tolist()
-    else:
-        strand_list = ["none"]
-        return strand_list * units.shape[0]
-
-
 rule count_matrix:
     input:
         expand(
@@ -14,21 +6,17 @@ rule count_matrix:
         ),
     output:
         "counts/all.tsv",
+    log:
+        "logs/count-matrix.log",
     params:
         samples=units["sample"].tolist(),
-        strand=get_strandness(units),
+        strand=get_strandedness(units),
     conda:
         "../envs/pandas.yaml"
     script:
         "../scripts/count-matrix.py"
 
 
-def get_deseq2_threads(wildcards=None):
-    # https://twitter.com/mikelove/status/918770188568363008
-    few_coeffs = False if wildcards is None else len(get_contrast(wildcards)) < 10
-    return 1 if len(samples) < 100 or few_coeffs else 6
-
-
 rule deseq2_init:
     input:
         counts="counts/all.tsv",
@@ -60,10 +48,6 @@ rule pca:
         "../scripts/plot-pca.R"
 
 
-def get_contrast(wildcards):
-    return config["diffexp"]["contrasts"][wildcards.contrast]
-
-
 rule deseq2:
     input:
         "deseq2/all.rds",

rules/qc.smk

@@ -26,7 +26,7 @@ rule rseqc_junction_annotation:
         "logs/rseqc/rseqc_junction_annotation/{sample}-{unit}.log",
     params:
         extra=r"-q 255",  # STAR uses 255 as a score for unique mappers
-        prefix="qc/rseqc/{sample}-{unit}.junctionanno",
+        prefix=lambda w, output: strip_suffix(output[0], ".junction.bed"),
     conda:
         "../envs/rseqc.yaml"
     shell:
@@ -45,7 +45,9 @@ rule rseqc_junction_saturation:
         "logs/rseqc/rseqc_junction_saturation/{sample}-{unit}.log",
     params:
         extra=r"-q 255",
-        prefix="qc/rseqc/{sample}-{unit}.junctionsat",
+        prefix=lambda w, output: strip_suffix(
+            output[0], ".junctionSaturation_plot.pdf"
+        ),
     conda:
         "../envs/rseqc.yaml"
     shell:
@@ -92,7 +94,7 @@ rule rseqc_innerdis:
     log:
         "logs/rseqc/rseqc_innerdis/{sample}-{unit}.log",
     params:
-        prefix="qc/rseqc/{sample}-{unit}.inner_distance_freq",
+        prefix=lambda w, output: strip_suffix(output[0], ".inner_distance.txt"),
     conda:
         "../envs/rseqc.yaml"
     shell:
@@ -123,7 +125,7 @@ rule rseqc_readdup:
     log:
         "logs/rseqc/rseqc_readdup/{sample}-{unit}.log",
     params:
-        prefix="qc/rseqc/{sample}-{unit}.readdup",
+        prefix=lambda w, output: strip_suffix(output[0], ".DupRate_plot.pdf"),
     conda:
         "../envs/rseqc.yaml"
     shell:
@@ -139,7 +141,7 @@ rule rseqc_readgc:
     log:
         "logs/rseqc/rseqc_readgc/{sample}-{unit}.log",
     params:
-        prefix="qc/rseqc/{sample}-{unit}.readgc",
+        prefix=lambda w, output: strip_suffix(output[0], ".GC_plot.pdf"),
     conda:
         "../envs/rseqc.yaml"
     shell:

rules/trim.smk

@@ -1,7 +1,3 @@
-def get_fastq(wildcards):
-    return units.loc[(wildcards.sample, wildcards.unit), ["fq1", "fq2"]].dropna()
-
-
 rule cutadapt_pe:
     input:
         get_fastq,

scripts/count-matrix.py

@@ -1,3 +1,7 @@
+import sys
+# logging
+sys.stderr = open(snakemake.log[0], "w")
+
 import pandas as pd
 
 def get_column(strandedness):