feat: Allow bam input files

.test/config/units.tsv

@@ -1,4 +1,4 @@
-sample	unit	fragment_len_mean	fragment_len_sd	fq1	fq2
+sample	unit	fragment_len_mean	fragment_len_sd	fq1	fq2	bam_single	bam_paired
 A	1			ngs-test-data/reads/a.chr21.1.fq	ngs-test-data/reads/a.chr21.2.fq
 B	1			ngs-test-data/reads/b.chr21.1.fq	ngs-test-data/reads/b.chr21.2.fq
 B	2	300	14	ngs-test-data/reads/b.chr21.1.fq	

.test/three_prime/config/units.tsv

@@ -1,4 +1,4 @@
-sample	unit	fragment_len_mean	fragment_len_sd	fq1	fq2
+sample	unit	fragment_len_mean	fragment_len_sd	fq1	fq2	bam_single	bam_paired
 SRR8309096	u1	430	43	quant_seq_test_data/SRR8309096.fastq.gz	
 SRR8309094	u1	430	43	quant_seq_test_data/SRR8309094.fastq.gz	
 SRR8309095	u1	430	43	quant_seq_test_data/SRR8309095.fastq.gz	

config/units.tsv

@@ -1,4 +1,4 @@
-sample	unit	fragment_len_mean	fragment_len_sd	fq1	fq2
+sample	unit	fragment_len_mean	fragment_len_sd	fq1	fq2	bam_single	bam_paired
 A	1			raw/a.chr21.1.fq	raw/a.chr21.2.fq
 B	1			raw/b.chr21.1.fq	raw/b.chr21.2.fq
 B	2	300	14	raw/b.chr21.1.fq	

workflow/rules/bam.smk

@@ -0,0 +1,27 @@
+rule separate_bam_paired:
+    input:
+        "{bam_file}.bam",
+    output:
+        "{bam_file}.1.fq",
+        "{bam_file}.2.fq",
+    log:
+        "{bam_file}.separate.log",
+    params:
+        fastq="-n",
+    threads: 3
+    wrapper:
+        "v3.10.2/bio/samtools/fastq/separate"
+
+
+rule samtools_fastq_interleaved:
+    input:
+        "{bam_file}.bam",
+    output:
+        "{bam_file}.fq",
+    log:
+        "{bam_file}.interleaved.log",
+    params:
+        " ",
+    threads: 3
+    wrapper:
+        "v3.10.2/bio/samtools/fastq/interleaved"

workflow/rules/common.smk

@@ -98,7 +98,20 @@ def is_single_end(sample, unit):
 
 def get_fastqs(wildcards):
     """Get raw FASTQ files from unit sheet."""
-    if is_single_end(wildcards.sample, wildcards.unit):
+    if not pd.isnull(units.loc[(wildcards.sample, wildcards.unit), "bam_single"]):
+        fqfrombam = units.loc[(wildcards.sample, wildcards.unit), "bam_single"].replace(
+            ".bam", ".fq"
+        )
+        return fqfrombam
+    elif not pd.isnull(units.loc[(wildcards.sample, wildcards.unit), "bam_paired"]):
+        fqfrombam1 = units.loc[
+            (wildcards.sample, wildcards.unit), "bam_paired"
+        ].replace(".bam", ".1.fq")
+        fqfrombam2 = units.loc[
+            (wildcards.sample, wildcards.unit), "bam_paired"
+        ].replace(".bam", ".2.fq")
+        return [fqfrombam1, fqfrombam2]
+    elif is_single_end(wildcards.sample, wildcards.unit):
         return units.loc[(wildcards.sample, wildcards.unit), "fq1"]
     else:
         u = units.loc[(wildcards.sample, wildcards.unit), ["fq1", "fq2"]].dropna()

workflow/rules/quant.smk

@@ -1,8 +1,10 @@
 rule kallisto_index:
     input:
-        fasta="resources/transcriptome.cdna.without_poly_a.fasta"
-        if is_3prime_experiment
-        else "resources/transcriptome.cdna.fasta",
+        fasta=(
+            "resources/transcriptome.cdna.without_poly_a.fasta"
+            if is_3prime_experiment
+            else "resources/transcriptome.cdna.fasta"
+        ),
     output:
         index="results/kallisto_cdna/transcripts.cdna.idx",
     log: