The overall goal of the Data Generation Aim is to produce high quality data measuring phenotypes, protein and phosphoproteomics, and RNA expression in 30 cell lines grown on diverse microenvironments (ME). This approach will significantly enhance the LINCS data matrix by providing phenotypic, proteomic, and transcriptomic data for cell lines under unique ME conditions. Furthermore, the data will be utilized for subsequent generation of cellular network signatures in the Data Analysis Aim.
Sub aim 1. 1 will be generate 10 different phenotypic endpoints for 30 cell lines grown on 3,060 MEs. We will utilize Microenvironment Microarrays (MEMA) to assess the effects of the ME on proliferation, apoptosis, differentiation state, cell binding, and motility in each cell line. Statistical analysis will identify 50 significant MEs for validation each year in years 2-6 in sub aim 1.2.A (total of 250 conditions). Each of the 50 ME conditions will be re-tested with the same endpoints at a second site using MEMA technology to provide independent validation of the primary results. At the same time, the cells will be tested using the 50 ME conditions on a second set of arrays designed to model the elastic modulus of human tissues to determine the effects on phenotypes of the matrix stiffness. From the 50 conditions tested in the validation aims, 30 of the most concordant between the primary and validation sites will be selected annually for analysis by Reverse Phase Protein Array (RPPA) aim 1.2.B and RNA expression analysis (aim I.2.C.). The RPPA aim will produce data using 500 validated antibodies against proteins and phosphoproteins in key signaling pathways under each of the ME perturbations. The RNA expression analysis will examine 1000 key genes using Luminex bead technology developed by the LINCS group at the Broad Institute, who will perform the same analysis for us using RNA prepared from each of the cell lines grown under the ME conditions. In total, the RPPA and RNA expression analysis will generate data for 150 different ME conditions under 2 different elastic moduli. All data will be carefully curated, with extensive metadata collected, and provided to the LINCS community for populating the data matrix.
The main goals of the data generation aim will be to populate the LINCS data matrix with unique information that is currently not present in the matrix, and to provide data for generation of cellular network signatures. Our proposal to use MEMA technology will identify unique combinatorial MEs that perturb various phenotypic endpoints, providing novel data to fill in the LINCS data matrix.