Concurrent MMBIRFinder

Recent literature has found Microhomology-mediated break-induced replication (MMBIR) to be a common mechanism producing complex mutations. Complex mutations underlie the genomic instability that results in various cancers. MMBIRFinder is a tool that has successfully detected the MMBIR mutations in budding yeast DNA by comparing sampled DNA reads to a whole reference genome. A major challenge for the MMBIRFinder is the heavy computational expense needed to find MMBIR regions in human DNA, where the size of a human DNA genome is very large compared to yeast DNA. In this work, we propose a modified approach, which utilizes parallel computations to significantly reduce the processing time consumed by the MMBIFinder to search for MMBIR regions in the human genome. Moreover, the newly proposed tool preserves the validity and accuracy of the original tool for yeast, where the original MMBIRFinder often fails to operate on the much larger human genome. This will allow for identification of complex mutations generated by MMBIR in human cancer genomes, thereby improving understanding of the roles played by MMBIR in generating the mutations and rearrangements that hallmark various types of cancers and genetic disease.