feat: add --chunk-size parameter to simgenotype and fix RuntimeError when reference coordinates extend past the final map file coordinate

docs/commands/simgenotype.rst

@@ -75,6 +75,10 @@ If speed is important, it's generally faster to use PGEN files than VCFs.
   --pop_field \
   --out tests/data/example_simgenotype.pgen
 
+.. warning::
+  Writing PGEN files will require more memory than writing VCFs. The memory will depend on the number of simulated samples and variants.
+  You can reduce the memory required by this step by writing the variants in chunks. Just specify a ``--chunk-size`` value.
+
 All files used in these examples are described :doc:`here </project_info/example_files>`.
 
 

haptools/__main__.py

@@ -202,6 +202,17 @@ def karyogram(bp, sample, out, title, centromeres, colors, verbosity):
         " continue to simulate a vcf file."
     ),
 )
+@click.option(
+    "-c",
+    "--chunk-size",
+    type=int,
+    default=None,
+    show_default="all variants",
+    help=(
+        "If requesting a PGEN output file, write genotypes in chunks of X variants; "
+        "reduces memory"
+    ),
+)
 @click.option(
     "-v",
     "--verbosity",
@@ -224,6 +235,7 @@ def simgenotype(
     sample_field,
     no_replacement,
     only_breakpoint,
+    chunk_size,
     verbosity,
 ):
     """
@@ -307,6 +319,7 @@ def simgenotype(
             no_replacement,
             out,
             log,
+            chunk_size,
         )
     end = time.time()
 

haptools/sim_genotype.py

@@ -23,8 +23,9 @@ def output_vcf(
         pop_field, 
         sample_field, 
         no_replacement,
-        out, 
-        log
+        out,
+        log,
+        chunk_size = None,
     ):
     """
     Takes in simulated breakpoints and uses reference files, vcf and sampleinfo, 
@@ -70,6 +71,8 @@ def output_vcf(
         output prefix
     log: log object
         Outputs messages to the appropriate channel.
+    chunk_size: int, optional
+        The max number of variants to write to a PGEN file together
     """
 
     log.info(f"Outputting file {out}")
@@ -215,7 +218,7 @@ def output_vcf(
             gts = GenotypesVCF(out, log=log)
 
     else:
-        gts = GenotypesPLINK(out, log=log)
+        gts = GenotypesPLINK(out, chunk_size=chunk_size, log=log)
 
     gts.samples = output_samples
     gts.variants = vcf.variants