ABSTRACT Stroke is the third leading cause of death in women and the leading cause of disability in the US. New proteomic techniques measure a wide profile of proteins, providing a comprehensive picture of protein functions. Since proteins perform most biological functions, they may be leveraged as biomarkers or drug therapy targets. Although results for proteomics and cardiovascular disease have been promising, only one small prospective proteomics study of ischemic stroke has been conducted. The over-arching goal of this study is to discover novel proteome-wide and multi-omic associations for incident ischemic stroke. Building on an established record of success in this R01, we will measure 3072 proteins (Olink platform) in the Nurses’ Health Studies (NHS; 460 ischemic strokes/ 460 controls) and VITamin D Omega3 TriAL (VITAL; 300 ischemic strokes/ 300 controls), as well as leverage separately measured data from the Women’s Health Initiative (WHI; 214 ischemic strokes), Framingham Heart Study Offspring Cohort (FHSOC; 103 ischemic strokes), and UK Biobank (UKB; ~450 ischemic strokes). In Aim 1, we will test candidate protein hypotheses in 1334 strokes in meta-analyses using Olink data from the NHS, VITAL, WHI, FHSOC and UKB. Ischemic stroke subtypes, and interactions by sex and race will be tested. In Aim 2, we will perform robust, proteome-wide discovery in NHS and VITAL with validation of significant proteins in meta- analyses of WHI, FHSOC, and UKB. Validated proteins will be evaluated for causal associations using mendelian randomization (MR) analyses. Using proteomic and genetic data from UKB (53,000), we will generate protein polygenic allele risk scores for the validated proteins and test these protein polygenetic risk scores for ischemic stroke in MEGASTROKE and SiGN. Ischemic stroke subtypes, and interactions by sex and race will be explored. In Aim 3, we will leverage prior measures of metabolomic profiles and extensive biomarker and risk factor data in NHS, VITAL and WHI, using novel integrative approaches, including similarity network fusion, to combine proteomic and metabolomic data, biomarker and risk factor data, in associations for ischemic stroke and stroke subtypes. Findings will be compared in NHS, VITAL and WHI. The proposed study leverages Olink proteomic data from five cohorts to identify proteins and multi-omic signatures integral to the incidence of stroke, which can be used to improve stroke prediction and prevention strategies and identify new drug targets for stroke prevention.