Only count high mapping quality (>50) supplementary reads - to avoid an issue in ONT/ultra-long ONT data where there are too many split reads. Supported gzipped file as input reference.
Just changed the version number to v1.3.1 to activate autobump from bioconda.