A python package to simulate structural variation Pysim-sv is a package for simulating high-throughput sequencing (HTS) data to evaluate perfor-mance of structural variation (SV) detection algorithms. Pysim-sv can introduce a wide spectrum of germline and somatic genomic variations, making simulated genomes more similar to real ge-nomes. The package contains functionalities to simulate aneuploidy as well as heterogeneous tumor data, which is very useful in assessing performance of algorithms in tumor studies. Further-more, Pysim-sv can introduce GC-biases, the most important and the most prevalent bias in HTS data, in the simulated HTS bias. Pysim-sv provides an unbiased toolkit for evaluating HTS-based SV detection algorithms.