Understanding evolution in cancer plays a critical role in prevention and cure of various types of cancer, such as colorectal cancer. A population of organisms evolve through speciation, genetic drift and natural selection, similarly tumors evolve by accumulation of mutations and the nature of these mutations (positive, negative, neutral). There are several factors which affect the evolution of cancer such as carcinogens and aging and the evolutionary mechanism behind these cancers can help us to better understand the overall architecture of a tumor and ultimately its cure. Ongoing tumor evolution can cause intratumoral heterogeneity (ITH) but there are spaciotemporal constraints restricting the growth of a subclone to a specific region in the tumor. There is a need to understand how a subclone grows from a primordial tumor and what is the overall architecture of the final tumor. Somatic mutations accumulated in the tumor faithfully dictate the evolutionary history of a clinically detected tumor. Sequencing data of bulk samples derived through next-generation sequencing (NGS) technology can help us to explore and analyze the nature of variants present in the tumor. However, challenges remain to reliably detect these somatic variants, especially at the low frequency range. Whereas a “golden” set of true low frequency mutations from bulk tumor sampling is lacking, we reason that mutations showing phylogenetic compatibility in multi-sample sequencing experiments are more likely to be real, as opposed to the ones that are incompatible. By studying how sequencing mapping features relate to phylogenetic compatibility, here we explore the utility of these features for a better filtration of low frequency somatic variants calls from bulk tumor sequencing.