Project Summary/Abstract. The goal of this BAATAAR-UP NCI ARTNet U54 application is to characterize and therapeutically counteract mechanisms of acquired resistance to molecularly-targeted therapies against mutant EGFR and KRAS in non-small cell lung cancer (NSCLC) by delineating the tumor-tumor microenvironment (TME) ecosystem and its plasticity during treatment. To achieve this goal, multi-omics data from annotated clinical specimens and several complimentary model systems will be generated. Bioinformatics, computational biology and biostatistics play an important role in this ARTNet Research Center. The major objective of the Data Science Core is to build and manage centralized multi-omics database and provide a full set of bioinformatical, computational and statistical support and integrate all 3 Project. This will include basic and translational science in systems such as clinical biopsies, PDX, PDO and cell line models, and integration of transcriptomics, spatial, genomics, proteomics and functional biology studies. We will contribute by providing computational and statistical support and applying and developing optimal bioinformatic, and statistical algorithms, tools and pipelines. The Core is staffed by expert faculty and computational scientists from the Bioinformatics and Computational Biology and Biostatistics Departments at MD Anderson and from the Bioengineering and Therapeutic Sciences Department at UCSF. This program’s PI and Co-investigators have previously worked closely and synergistically with Data Science Core’s investigators in other projects and grant applications. The Data Science Core will work closely with Projects 1–3 and the Administrative Core to manage and analyze the data resources utilizing the existing, robust IT structure in place at MD Anderson and UCSF. The Data Science Core has built various pipelines and algorithms for “-omics” and functional biology data processing and analyses. The Core will apply these pipelines and algorithms to all types of data generated. The Core will utilize standard design principles, bioinformatical, computational and statistical algorithms, and will develop new methods as needed to analyze all data collected in these projects, including spatial transcriptomics, cell-cell interaction analysis, and CRISPR- and proteomic profiling. Parametric and nonparametric methods will be used for parameter estimation and hypothesis testing. Linear models and generalized additive models will be used to find the best models to fit complex data structures. The Core will facilitate hypothesis testing across projects by integrating datasets from multiple laboratories using various algorithms, including Bayesian network-based models and Modular Analysis of Genomic NET works In Cancer (MAGNETIC). All data analyses will be performed using R and Bioconductor packages. The Core will document all analyses and produce HTML or PDF reports (using R packages: Sweave, knitR, markdown) for documentation and reprod...