Abstract: Globally, about 2 million people are diagnosed with lung cancer annually, making it the most common type of cancer in the world. In addition, lung cancer is the leading cause of cancer death among both men and women, accounting for \textasciitilde 25% of all cancer deaths in the United States and \textasciitilde 19% of cancer deaths worldwide. Although tobacco smoking is the most common cause of lung cancer, it is estimated that 10-25% of all patients diagnosed with lung cancer worldwide are never smokers, with higher proportions in Asians and women. Compared to former and current smokers with lung cancer, lung cancers in never smokers tend to present at an earlier age and more frequently have adenocarcinoma histology. Although a few risk factors are known to contribute to the etiology of lung cancer in never smokers, a large fraction of cancer cases cannot be explained by established environmental and genetic risk factors, highlighting the need for research in this area. One promising approach to identify the etiological factors involved in lung tumorigenesis in never smokers is based on the study of the “mutational signatures” that the exogenous and endogenous processes leave on the tumor tissue and surrounding areas. Somatic mutations accumulate throughout individuals’ lifetime, as a result of the balance between DNA damage and DNA repair. Mutational sources can change over time and mutational signatures can provide information on the tumor evolutionary trajectory, tracking activities of mutational processes. We designed “Sherlock-Lung” to identify mutational signatures in lung cancers from never smokers and order them based on their clonal and subclonal localization to have an insight on the factors involved in the initiation of the tumors. Prevention and treatment strategies are likely to be more effective when targeting the tumor initiating factors rather than those responsible for the tumor progression. Here, we present the Sherlock-Lung study design, which includes a combination of lung cancer cases from populations highly exposed to lung cancer risk factors and populations with unknown exposures. We also present preliminary data from this study and what we have learned from whole genome sequencing analysis of the mutational landscape of lung cancers from smokers and nonsmokers and from different histological subtypes. For example, although the mutational signature associated with tobacco smoking is largely clonal in all lung cancers from smokers, consistent with an initiating role of tobacco smoking in lung tumorigenesis, the mutational signatures distribution varies across histological subtypes. Moreover, the frequency of mutations in major cancer driver genes varies dramatically between smokers and never smokers, and mutational signatures distribution appears to differ between tumors with and without specific driver genes. In never smokers, without the dominant mutational source of tobacco smoking, other signatures emerged, e.g., that associated with reactive oxygen species. Although this approach has already revealed important information on the etiological processes involved in tumorigenesis of lung cancer and other cancer types, there is a growing list of open questions and challenges that need to be addressed. In population studies, careful exposure assessment to identify and validate new mutational signatures and the inclusion of study subjects from different geographical locations and ethnicities are critical.