Bayesian approach and Imprecise Dirichlet Model for the selection of (underused/overused) medicinal plants in the Guinean zone of Togo

Abstract

Identifying priority medicinal plants at the local level is crucial for the improvement of integrative and strong conservation solutions around these resources. This study was carried out to determine underused and overused botanical families, and, analyse the correlation between botanical families, used plant parts, and categories of treated diseases in the Guinean zone of Togo. Ethnobotanical field surveys conducted via semi-structured interviews (individuals/focus groups) among traditional healers and medicinal plant sellers helped to identify the rich diversity of medicinal plants used in the Guinean zone. Bayesian statistics and the Imprecise Dirichlet Model (IDM) were used to detect botanical families' level of use (under/overused). A matrix analysis via phylogeny coupled with a cladogram was carried out to highlight the preferences of users in plant parts and botanical families in the treatment of diseases. It appears that 12.43% of the local flora in the Guinean zone of Togo is used in traditional medicine. The most treated diseases with a high plant diversity are general and unspecified categories of diseases (228 plants), followed by cardiovascular diseases (133 plants). Both Bayesian and IDM approaches show that the most overused families are Amaranthaceae, followed by Combretaceae, Euphorbiaceae, Meliaceae, Myrtaceae, and Rutaceae. The underused families are Cyperaceae, Orchidaceae, and Poaceae. The leaves and roots are the plant parts in high demand for the preparation of medicinal drugs. The presence of chemical groups such as phenols, flavonoids, tannins, saponosides, coumarins, and alkaloids would justify the choice of a botanical family or specific plant parts for the treatment of a category of diseases. It is necessary to develop conservation strategies for overused plant families to ensure sustainability for future generations.