Updated project metadata. Background: Heart failure with preserved ejection fraction (HFpEF) constitutes more than half of all heart failure but has few effective therapies. Recent human myocardial transcriptomics and metabolomics have revealed major differences between HFpEF, HF with reduced EF (HFrEF), and controls. How this translates at the protein level is currently unknown. Methods: Myocardial tissue from patients with HFpEF and non-failing donor controls was analyzed by data-dependent (DDA, n=10 HFpEF, n=9 controls) and data-independent (DIA, n=44 HFpEF, n=5 controls) mass spectrometry-based proteomics. Previously reported myocardial proteomic data from end-stage HFrEF and controls were also used. Differential protein expression analysis, machine learning and pathway enrichment were integrated with clinical characteristics and myocardial transcriptomics. Results: DDA-MS proteomics identified 88 significantly upregulated and 248 down-regulated proteins in HFpEF vs controls, out of 1996 identified proteins. Principal component analysis of DDA-MS proteomics found HFpEF was separated into 2 sub-groups: one being similar to controls the other quite disparate. Top proteins contributing to the separation of HFpEF subgroups were enriched in actin/myosin binding, regulation of DNA replication/repair, transcription, and translation. Downregulated proteins in HFpEF vs controls were enriched in pathways related to ribosome structure, transmembrane transporters, metabolic enzymes, and oxidative phosphorylation (OxPhos) proteins. Enriched pathways for proteins upregulated in HFpEF related to actin and phospholipid binding, growth factor signaling, kinase regulation, and glycolysis. Ingenuity pathway analysis predicted downregulation of protein translation, mitochondrial function, and glucose and fat metabolism in HFpEF. OxPhos gene (increased) versus protein (decreased) expression was discordant in HFpEF. The second DIA proteomic analysis also yielded two HFpEF sub-groups; the one most different from controls also having reduced OxPhos and protein translation pathways. A higher proportion of these patients also had severe obesity. Conclusions: Integrative proteomics, transcriptomics, and pathway analysis supports a translational defect particularly involving mitochondrial, ribosomal and protein translation proteins in HFpEF. Patients with more distinct proteomic signatures from control were more often very obese. The results support therapeutic efforts targeting metabolism, mitochondrial function, and protein translation in this subgroup.