Check genome size param for span=auto (#57)

* Set G=0 by default and check value in tigmint-make

* Bump up version numbers to 1.2.4

* Fix typo

* Remove and gitignore compiled long-to-linked-pe

Author: janetxinli

.gitignore

@@ -26,6 +26,7 @@ xml-patch-make/
 __pycache__
 .pytest_cache
 .vscode
+src/long-to-linked-pe
 
 # BWA
 *.fa.amb

README.md

@@ -149,7 +149,7 @@ tigmint-make tigmint-long draft=myassembly reads=myreads span=auto G=gsize dist=
 
 + `draft`: Name of the draft assembly, `myassembly.fa`
 + `reads`: Name of the reads, `myreads.fq.gz`
-+ `G`: Haploid genome size of the draft assembly organism. Used to calculate `span` parameter automatically. Can be given as an integer or in scientific notation (e.g. '3e9' for human)
++ `G`: Haploid genome size of the draft assembly organism. Required to calculate `span` parameter automatically. Can be given as an integer or in scientific notation (e.g. '3e9' for human) [0]
 + `span=20`: Number of spanning molecules threshold. Set `span=auto` to automatically select span parameter (currently only recommended for `tigmint-long`)
 + `cut=500`: Cut length for long reads (`tigmint-long` only)
 + `longmap=ont`: Long read platform; `ont` for Oxford Nanopore Technologies (ONT) long reads, `pb` for PacBio long reads (`tigmint-long` only)

bin/tigmint-cut

@@ -242,7 +242,7 @@ def get_span(filename):
 
 def main():
     parser = argparse.ArgumentParser(description="Find misassembled regions in assembly using Chromium molecule extents")
-    parser.add_argument("--version", action="version", version="tigmint-cut 1.2.3")
+    parser.add_argument("--version", action="version", version="tigmint-cut 1.2.4")
     parser.add_argument("fasta", type=str, help="Reference genome fasta file (must have FAI index generated)")
     parser.add_argument("bed", type=str, help="Sorted bed file of molecule extents")
     parser.add_argument("-o", "--fastaout", type=str, help="The output FASTA file.", required=True)

bin/tigmint-make

@@ -33,7 +33,7 @@ endif
 ref=ref
 
 # Haploid size of the reference genome, for calculating span, NG50 and NGA50
-G=-1
+G=0
 
 # Minimap2 long read map parameter
 longmap=ont
@@ -114,7 +114,7 @@ endif
 
 .DELETE_ON_ERROR:
 .SECONDARY:
-.PHONY: help version all tigmint tigmint-long arcs metrics draft_metrics tigmint_metrics arcs_metrics
+.PHONY: help version all tigmint tigmint-long arcs metrics draft_metrics tigmint_metrics arcs_metrics check_span_g
 
 help:
 	@echo 'Tigmint: Correct misassemblies using linked or long reads'
@@ -123,12 +123,12 @@ help:
 	@echo 'For more information see https://bcgsc.github.io/tigmint/'
 
 version:
-	@echo "Tigmint 1.2.3"
+	@echo "Tigmint 1.2.4"
 	@echo "Written by Shaun Jackman @sjackman."
 
 all: tigmint arcs
 ifneq ($(ref), ref)
-ifneq ($(G), -1)
+ifneq ($G, 0)
 all: metrics
 endif
 endif
@@ -196,19 +196,27 @@ $(draft).%.sortbx.bam: %.fq.gz $(draft).fa.bwt
 $(draft).%.cut$(cut).sortbx.bam: %.cut$(cut).fa.gz $(draft).fa
 	$(gtime) minimap2 -y -t$t -ax map-$(longmap) --secondary=no $(draft).fa $< | samtools view -b -u -F4 | samtools sort -@$t -tBX -T$$(mktemp -u -t $@.XXXXXX) -o $@
 
+# Check that G is set if span=auto
+check_span_g:
+ifeq ($(span), auto)
+ifeq ($G, 0)
+	$(error Must set genome size parameter (G) to calculate span automatically)
+endif
+endif
+
 # Segment long reads from gzipped fasta file, optionally calculating tigmint-long parameters.
-$(reads).cut$(cut).fq.gz: $(longreads)
+$(reads).cut$(cut).fq.gz: $(longreads) check_span_g
 ifeq ($(span), auto)
 ifeq ($(dist), auto)
-	$(gtime) $(gzip) -dc $< | $(bin)/long-to-linked -l$(cut) -m$(minsize) -g$(G) -s -d -o $(reads).tigmint-long.params.tsv | $(gzip) > $@
+	$(gtime) $(gzip) -dc $(longreads) | $(bin)/long-to-linked -l$(cut) -m$(minsize) -g$G -s -d -o $(reads).tigmint-long.params.tsv | $(gzip) > $@
 else
-	$(gtime) $(gzip) -dc $< | $(bin)/long-to-linked -l$(cut) -m$(minsize) -g$(G) -s -o $(reads).tigmint-long.params.tsv | $(gzip) > $@
+	$(gtime) $(gzip) -dc $(longreads) | $(bin)/long-to-linked -l$(cut) -m$(minsize) -g$G -s -o $(reads).tigmint-long.params.tsv | $(gzip) > $@
 endif
 else
 ifeq ($(dist), auto)
-	$(gtime) $(gzip) -dc $< | $(bin)/long-to-linked -l$(cut) -m$(minsize) -d -o $(reads).tigmint-long.params.tsv | $(gzip) > $@
+	$(gtime) $(gzip) -dc $(longreads) | $(bin)/long-to-linked -l$(cut) -m$(minsize) -d -o $(reads).tigmint-long.params.tsv | $(gzip) > $@
 else
-	$(gtime) $(gzip) -dc $< | $(bin)/long-to-linked -l$(cut) -m$(minsize) | $(gzip) > $@
+	$(gtime) $(gzip) -dc $(longreads) | $(bin)/long-to-linked -l$(cut) -m$(minsize) | $(gzip) > $@
 endif
 endif
 
@@ -229,7 +237,7 @@ $(reads).tigmint-long.params.tsv: $(longreads)
 	$(bin)/tigmint_estimate_dist.py - -n $(dist_sample) -o $@'
 
 # Create molecule extents BED using cut long reads
-$(draft).$(reads).cut$(cut).molecule.size$(minsize).bed: $(longreads) $(draft).fa $(reads).tigmint-long.params.tsv
+$(draft).$(reads).cut$(cut).molecule.size$(minsize).bed: $(longreads) $(draft).fa $(reads).tigmint-long.params.tsv check_span_g
 ifeq ($(dist), auto)
 	$(gtime) $(bin)/../src/long-to-linked-pe -l $(cut) -m$(minsize) -g$G -s -b $(reads).barcode-multiplicity.tsv --bx -t$t --fasta -f $(reads).tigmint-long.params.tsv $< | \
 	minimap2 -y -t$t -x map-$(longmap) --secondary=no $(draft).fa - | \
@@ -326,7 +334,7 @@ endif
 %.abyss-fac.tsv: %.fa
 	abyss-fac -G$G -t500 $< >$@
 
-ifneq ($(G), -1)
+ifneq ($G, 0)
 abyss_samtobreak=abyss-samtobreak -l500 -G$G
 else
 abyss_samtobreak=abyss-samtobreak -l500

bin/tigmint_estimate_dist.py

@@ -44,7 +44,7 @@ def main():
                         required=False)
     parser.add_argument("-v", "--version",
                         action="version",
-                        version="tigmint_estimate_dist.py 1.2.3")
+                        version="tigmint_estimate_dist.py 1.2.4")
 
     args = parser.parse_args()
 

bin/tigmint_molecule.py

@@ -243,7 +243,7 @@ def parse_arguments(self):
             "Read a SAM/BAM file and output a TSV file. "
             "The SAM/BAM file must be sorted by BX tag and then by position.")
         parser.add_argument(
-            '--version', action='version', version='tigmint-molecule 1.2.3')
+            '--version', action='version', version='tigmint-molecule 1.2.4')
         parser.add_argument(
             metavar="BAM", dest="in_bam_filename",
             help="Input BAM file sorted by BX tag then position, - for stdin")

bin/tigmint_molecule_paf.py

@@ -102,7 +102,7 @@ def parse_arguments(self):
         parser = argparse.ArgumentParser(
             description="Group linked reads simulated from long reads into molecules. "
                         "Read a PAF file and output a BED file.")
-        parser.add_argument('--version', action='version', version='tigmint_molecule_paf.py 1.2.3')
+        parser.add_argument('--version', action='version', version='tigmint_molecule_paf.py 1.2.4')
         parser.add_argument(metavar="PAF", dest="PAF", help="Input PAF file, - for stdin")
         parser.add_argument("-o", "--output", dest="out_molecules_filename",
                             help="Output molecule BED file [stdout]",

setup.py

@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name="tigmint",
-    version="1.2.3",
+    version="1.2.4",
     author="Shaun Jackman",
     author_email="[email protected]",
     description="Correct misassemblies using linked or long reads",

src/long-to-linked-pe.cpp

@@ -21,7 +21,7 @@
 #include <vector>
 
 const static std::string PROGNAME = "long-to-linked-pe";
-const static std::string VERSION = "v1.2.3";
+const static std::string VERSION = "v1.2.4";
 const static size_t MAX_THREADS = 6;
 
 static void