Multi-Source Transcriptomic and Structural Prioritization of Adansonia digitata Phytochemicals in Myocardial Infarction

Background/Objectives: Myocardial infarction (MI) remains a major therapeutic challenge, and the molecular targets potentially associated with the cardiovascular activity of Adansonia digitata phytochemicals remain incompletely characterized. This study integrated phytochemical, transcriptomic, disease-gene, network, and structural evidence to prioritize candidate phytochemical–MI target associations. Methods: Ten polyphenols were selected from published LC-MS/QTOF data. Targets predicted using SwissTargetPrediction and PharmMapper were integrated with MI-associated evidence from the GSE66360 (circulating endothelial cell), GSE325257 (cardiac targeted gene-expression), and DisGeNET. An initial composite analysis generated a 19-gene discovery module for network and structural prioritization. A complementary source-resolved sensitivity analysis independently evaluated the three MI evidence sources. Selected interactions were examined by molecular docking, 200 ns molecular dynamics simulations, and MM/PBSA binding free-energy estimation. Results: Sensitivity analysis identified 189 genes supported by at least two MI evidence sources, including 21 predicted Adansonia digitata targets. Four genes (GSK3B, NOS3, SOD2, and TGFBR1) were shared between the 19-gene discovery and 21-gene sensitivity modules. Gene Ontology analysis identified 36 significantly overrepresented biological-process terms (FDR < 0.05). Docking supported the selection of chlorogenic acid and rutin for simulation with MMP2 and NOS3. MM/PBSA calculations indicated that chlorogenic acid exhibited more favorable estimated binding energies compared to rutin, especially in the case of MMP2. However, these differences were not statistically validated. Conclusions: The complementary integration strategies prioritized candidate Adansonia digitata phytochemicals and MI-associated targets and generated testable hypotheses for further investigation. These computational findings are hypothesis-generating and require experimental validation of target engagement, biological activity, pharmacokinetics, and efficacy in MI models.