**AWARDS ISSUED PRIOR TO JANUARY 20, 2025, WERE FUNDED UNDER PREVIOUS ADMINISTRATIONS AND MAY NOT REFLECT THE PRIORITIES AND POLICIES OF THE CURRENT ADMINISTRATION.** GENETIC MAPPING LETS BIOLOGISTS DETERMINE THE GENETIC ARCHITECTURE OF LARGE-EFFECT VARIANTS ON TRAITS OF INTEREST (E.G. QTLS). HOWEVER, IT REMAINS A PIECEMEAL PROCESS FOR BREEDERS TO LEVERAGE QTLS TO IDENTIFY ASSAYABLE MARKERS FOR MARKER-ASSISTED SELECTION AND GENERATE CANDIDATE GENES FOR MODIFICATION. WITH THE ADVENT OF MULTIPLE REFERENCE GENOMES, DIVERSITY DATA, AND HIGH-CONFIDENCE GENE ANNOTATIONS, ALL OF THE RAW MATERIALS ARE AVAILABLE TO IDENTIFY HIGH-VALUE MOLECULAR TARGETS FOR BREEDING. HOWEVER, A COMPUTATIONAL FRAMEWORK TO MAKE USE OF THESE DATA DOES NOT EXIST YET FOR COTTON. THE WORK PROPOSED HERE WILL DEVELOP AN EXTENSIBLE AND ACCESSIBLE FRAMEWORK TO LEVERAGE MULTIPLE REFERENCE GENOMES OF COTTON TO EFFECTIVELY FIND TARGETS FOR MOLECULAR BREEDING EFFORTS.HERE, WE WILL BUILD AND VETT TWO RESOURCES TO ACCOMPLISH THIS GOAL. FIRST, MULTIPLE GENOME SEQUENCES WILL BE INTEGRATED INTO A SINGLE SYNTENY-CONSTRAINED 'PAN-GENOME' THAT CORRECTLY HANDLES THE TWO SUBGENOMES OF TETRAPLOID COTTON AND ANCHORS GENES TO THE PHYSICAL COORDINATES OF MULTIPLE REFERENCE GENOMES. SECOND, WE WILL FUNCTIONALLY ANNOTATE THE PAN-GENOME WITH VARIANTS ACROSS THE DIVERSITY OF COTTON AND BUILD A COMPUTATIONAL FRAMEWORK (AND USER-FACING QUERY TOOLS) TO TRANSFORM VARIANTS INTO RANKED SETS OF CANDIDATE GENES. THESE RESOURCES WILL ALLOW RESEARCHERS TO INTERROGATE PHYSICAL REGIONS OF INTEREST (E.G. QTLS), WHICH WILL FORM A STRONG FOUNDATION FOR MOLECULAR BREEDING AND IMPROVEMENT OF COTTON GERMPLASM. THE PROPOSAL IS SUBMITTED TO FY2021 AFRI COMPETITIVE GRANTS PROGRAM: FOUNDATIONAL AND APPLIED SCIENCE PROGRAM PRIORITY AREA: INITIATE INCREMENTAL DEVELOPMENT OF A PANGENOME AND HAPLOTYPE DATABASE FOR GOSSYPIUM HIRSUTUM UPLAND COTTON BREEDING, CO-FUNDED BY THE COTTON BOARD.