Publicação
Development of a biomarker prioritisation algorithm for precision medicine
| datacite.subject.fos | Ciências Médicas::Ciências da Saúde | |
| datacite.subject.sdg | 09:Indústria, Inovação e Infraestruturas | |
| datacite.subject.sdg | 03:Saúde de Qualidade | |
| dc.contributor.advisor | Rosa, Nuno Ricardo das Neves | |
| dc.contributor.advisor | Duarte, José Alexandre Graça | |
| dc.contributor.author | Kaddouri, Walid | |
| dc.date.accessioned | 2026-08-07T16:34:29Z | |
| dc.date.available | 2026-08-07T16:34:29Z | |
| dc.date.embargo | 2027-08-07 | |
| dc.date.issued | 2026-07-23 | |
| dc.date.submitted | 2026-01-01 | |
| dc.description.abstract | Precision medicine is gradually shifting clinical reasoning away from population-level averages and toward the molecular profile of the individual patient . Salivary biomarkers are well suited to this shift, since saliva can be sampled non-invasively, repeatedly and at low cost, and contains more than three thousand identifiable proteins reflecting both oral and systemic biology. The bottleneck is not the amount of available data. The salivary proteomic literature has grown considerably over the past two decades, yet no clear methodological framework exists to select, among the thousands of detectable proteins, those that genuinely deserve experimental validation as candidate biomarkers. Machine-learning approaches partially address this question but remain difficult to interpret in clinical practice. SalivatecSmartBiomarkers was developed in response to this gap. It is an open web platform that gathers 3,253 proteomic entries across sixteen disease contexts and assigns each protein a composite score built from three independent biological dimensions: the amplitude of expression variation between diseased and healthy states (fold change), the protein’s coverage of human biological pathways via Reactome, and its gene-disease associations via DisGeNET , with disease relevance assessed through a ternary d1/d2/d3 classification based on MeSH semantic similarity. The score runs from −10 to 100 points, and each point is tied to an explicit, traceable biological criterion. The algorithm was validated on chronic periodontitis using 1,388 proteomic observations. A Spearman correlation of rs = 0.52 (p = 0.003) was obtained between composite scores and an independent bibliographic validation. Calgranulin B (S100-A9) reaches the maximum on both validation tiers (BVS-1 = 10/10, BVS-2 = 100/100), which is consistent with its known periodontal biology: a neutrophilic alarmin released at sites of active inflammation, it sequesters the divalent metals required by bacterial enzymes and amplifies the inflammatory response through TLR4 and RAGE. This work proposes a formal rule-based composite scoring framework dedicated to the prioritization of salivary biomarkers in precision medicine. | eng |
| dc.identifier.other | e40f2d55-5aed-4f07-a43a-017cc227c342 | |
| dc.identifier.tid | 204341256 | |
| dc.identifier.uri | http://hdl.handle.net/10400.14/59089 | |
| dc.language.iso | eng | |
| dc.rights.uri | N/A | |
| dc.subject | Saliva | eng |
| dc.subject | Biomarkers | eng |
| dc.subject | Periodontitis | eng |
| dc.title | Development of a biomarker prioritisation algorithm for precision medicine | por |
| dc.type | master thesis | |
| dspace.entity.type | Publication | |
| thesis.degree.name | Mestrado em Medicina Dentária |
Ficheiros
Principais
1 - 1 de 1
Miniatura indisponível
- Nome:
- 204341256.pdf
- Tamanho:
- 1.22 MB
- Formato:
- Adobe Portable Document Format