Individual tumors were dissected, digested in an enzymatic buffer (1X HBSS, 5mM HEPES, DNaseI, Collagenase IV), and incubated with rotation at 37°C for 30 minutes. The enzymatic buffer was quenched with DMEM and spun at 1000 rpm. Cell pellets were resuspended in DMEM and plated in 6-well plates to allow for attachment. Cell lines were genotyped for Kras, p53, and tomato after 5 passages in culture. The cell lines used in this study were established from mouse LUAD over the course of the study. All lines were grown in DMEM, 10% FBS, and 1% pen-strep. KP cell lines have not been authenticated because the cell lines are not found in established databases. The KP cell lines were tested for mycoplasma and found to be negative. GM12878 cells were grown in DMEM, 10% FBS, and 1% pen-strep and 3T3 cells were grown in RPMI 1640, 10% FBS, and 1% pen-strep. GM12878 cells were authenticated by STR Profiling Service from ATCC. For CRISPR knockout experiments, guides were cloned into the lentiCRISPR-V2 lentiviral vector (Joung et al., 2017). The lentiCRISPR-V2 vector was digested with Fast Digest EspI and ligated with EspI-compatible annealed oligonucleotides for sgRNAs. KP cell lines were infected with constructs containing guides and selected with puromycin after 48 hours. After puromycin selection, guide performance was tested by western blotting. For CRISPR activation (CRISPRa) experiments, the Lenti-Sam-puro construct was used (developed in the Jacks lab), an adaption of the previously published Lenti-Sam activation construct (Pentinmikko et al., 2019). EspI-compatible cloning was completed and cells were infected with constructs containing guides based on CRISPRa prediction software. Truncated guides of 15 bp were cloned into a lentiviral based expression construct which also encodes for a transcriptional activation complex (MS2-P65-HSF1) and a puromycin selection cassette. Non-metastatic and metastatic cell lines were engineered to express Cas9 following stable selection of a Cas9-Blast construct.