Natural Language Processing of Biomedical Text to Map and Prioritize Protein–Disease Associations in HFpEF
Computers in Biology and Medicine, 205: 111599 · DOI: 10.1016/j.compbiomed.2026.111599
The Problem
Identifying clinical biomarkers and drug targets for cardiovascular disease has traditionally relied on biochemical and omics experiments. These studies have generated millions of publications — but the knowledge they contain is scattered across a fragmented, ever-growing literature that now exceeds 38 million indexed papers on PubMed alone.
No single researcher, lab, or even institution can synthesise that body of work. Proteins of high diagnostic or therapeutic relevance go unnoticed simply because the evidence for them is buried across thousands of papers that have never been read side-by-side.
This problem is especially acute for heart failure with preserved ejection fraction (HFpEF), a subtype of heart failure where the left ventricle contracts normally but stiffens during filling. HFpEF is entangled with aging, obesity, diabetes, hypertension, and hyperlipidemia, has no approved disease-modifying therapy, and remains poorly understood at the molecular level.
Previous text-mining efforts in cardiology have focused on electronic health records or narrow bibliometric analyses. No study had previously applied NLP at this scale to systematically map protein–disease associations across all major CVD groups and then focused specifically on HFpEF.
The Approach
The authors developed VITAL (ValIdated Text-mining using Advanced Language models), a computational framework built around the CaseOLAP algorithm (Context Aware Semantic Online Analytical Processing). VITAL scores each of the 20,428 known human proteins against every disease domain of interest, producing a ranked list of protein–disease associations derived entirely from published text.
-
Step 01
Corpus assembly via MeSH descriptors
All PubMed abstracts relevant to six major CVD groups (ischemic heart disease, cardiomyopathy, cerebrovascular accident, arrhythmia, valvular disease, and congenital heart disease) plus HFpEF and its comorbidities were retrieved using Medical Subject Headings, yielding 5.5 million curated abstracts from 38 million total publications.
-
Step 02
CaseOLAP scoring — integrity, distinctiveness, popularity
For each protein–disease pair, the algorithm computes three quantities: how reliably the protein name refers to a single concept (integrity), how exclusively the protein appears in that disease's literature versus other diseases (distinctiveness), and how frequently it is mentioned in that disease domain (popularity). Their product is the final attribution score.
-
Step 03
HFpEF deep dive — comorbidities and pathological mechanisms
Using HFpEF as the primary case study, the framework mapped protein scores across four comorbidity domains (aging, type 2 diabetes/obesity, hypertension, hyperlipidemia) and three pathological mechanisms (mitochondrial dysfunction, inflammation, fibrosis), identifying proteins shared and unique across each domain.
-
Step 04
Validation against real patient biopsy data
The top computationally ranked HFpEF proteins were cross-referenced against a myocardial proteomics dataset from HFpEF patients and age-matched controls. A hypergeometric test confirmed that the overlap between computational predictions and biopsy-detected proteins was 2.65× greater than expected by chance (p = 0.00265).
Key Findings
Applied to the full PubMed corpus, VITAL mapped protein–disease associations across six CVD groups and produced a rich molecular portrait of HFpEF and its biological neighbourhood.
IHD, CM, and CVA share the densest molecular landscapes
Ischemic heart disease, cardiomyopathy, and cerebrovascular accidents each had over 7,000 associated proteins and substantial pairwise overlap — CM and IHD alone share 6,595 proteins — suggesting convergent molecular pathways across these three conditions.
Known biomarkers confirmed at high scores
Established clinical biomarkers — troponin-I (IHD), BNP (cardiomyopathy), ACE and renin (heart failure), TNF and galectin-3 (inflammation) — all received CaseOLAP scores above the statistically derived cut-off for their respective diseases, validating the scoring framework.
4,879 proteins converge across HFpEF and all comorbidities
Of 5,124 proteins found in HFpEF literature, 4,879 are shared across all four comorbidity domains. Aging showed the strongest computational correlation with HFpEF (r = 0.61), suggesting the deepest shared biological pathways of any comorbidity studied.
Eight top-ranked proteins are consistent across all domains
TDO2, SPARC, TCO1, MYPC3, and ASA were the highest-ranked HFpEF proteins and remained top-ranked across comorbidities and pathological mechanisms alike, implying they are central molecular drivers rather than domain-specific signals.
HFpEF proteins localise outside the heart
Cross-referencing with the Human Protein Atlas revealed that the top computationally ranked proteins are primarily expressed in liver, pancreas, and adipose tissue — not cardiac tissue — providing computational evidence that HFpEF is fundamentally a multisystemic disorder driven by inter-organ signalling.
Computational predictions validated in patient biopsies
Seven of the top 20 NLP-ranked HFpEF proteins appeared in the top 100 differentially expressed proteins in myocardial biopsies from HFpEF patients, with an enrichment factor of 2.65× (p = 0.00265) over random expectation. Vinculin, desmin, and SPARC were among the validated candidates.
Why It Matters
VITAL is not a replacement for wet-lab science — it is a navigation system for it. By ranking proteins against disease domains at the scale of the entire biomedical literature, it surfaces candidates that experiments alone would take decades to reach.
- Biomarker Discovery Proteins that score high in exclusivity and frequency for a given disease are strong candidates for diagnostic or prognostic biomarkers. VITAL systematically surfaces these candidates from millions of publications, complementing and guiding costly experimental omics studies.
- Drug Target Identification High-scoring proteins with known molecular functions (like TDO2 in the microbiome–heart axis, or SPARC in cardiac remodelling) represent actionable targets for therapeutic development, especially in HFpEF where no approved disease-modifying therapy currently exists.
- Systems Medicine By mapping 4,991 proteins shared across HFpEF and its three key pathological mechanisms, the framework reveals convergent molecular networks that operate across inflammation, mitochondrial dysfunction, and fibrosis simultaneously, pointing to multi-target intervention strategies.
- HFpEF as a Systemic Disease The finding that top-ranked HFpEF proteins are predominantly expressed in liver, pancreas, and adipose tissue rather than heart tissue provides computational corroboration for the emerging clinical view that HFpEF is driven by inter-organ crosstalk and metabolic dysregulation.
- Scalable to Other Diseases VITAL requires only a text corpus and a protein list. The same pipeline can be applied to any disease with a PubMed presence, making it a general-purpose framework for protein prioritisation across the full spectrum of human disease.
Infographics
Paper & Authors
Authors
Cite This Work
Open access · CC BY 4.0