Cancer development and progression is driven by genetic alterations. These alterations include somatic DNA sequence changes (somatic SNVs), copy number aberrations (CNAs), chromosomal rearrangements, and epigenetic modification. For over 20 years scientists have been studying cancer genetics to develop new and better treatment options. Within the past few years cancer genetic studies have shifted from single gene studies to whole genome studies with the development of next- generation sequencing (NGS) platforms. One challenge with using NGS platforms to identify somatic mutations in a tumor is determining which somatic mutations are true positives vs. false positives. Here we developed different methodologies to remove sequencing errors caused by various sample preparations and sequencing platforms in order to identify true functional somatic variants in clinical samples. We analyzed and evaluated cancer specimens processed in FFPE, pre-clinical tumor models, tumors with low tumor cellularity, and tumor subclones with the use of next generation sequencing data. First, we analyzed whole-genome sequencing data from formalin-fixed paraffin embedded breast cancer samples and developed a new method for filtering false positive somatic mutations caused by formalin damaging the DNA. Second, we evaluated the genomic validity of tumor derived pre-clinical models using exome sequencing data. Finally, we developed a new software suite called Mutascope that is designed to analyze PCR amplicon sequencing data and identify somatic variants that are present in only 1% of the tumor