UVM Theses and Dissertations
Format:
Print
Author:
Sharma, Vivekanand
Dept./Program:
Microbiology and Molecular Genetics
Year:
2014
Degree:
Ph. D.
Abstract:
A vast magnitude of information is documented regarding the medicinal use of plants ranging from ancient herbal use to contemporary biomedical and clinical investigations. The medicinal importance of plants ("phyto-therapies") using contemporary scientific techniques and results are incorporated into databases or published as texts complements ongoing efforts to digitize previous records of plant-based medicinal uses. The current data regarding phyto-therapies are spread across many databases in different formats. In additionl plant-specific data are embedded within the realm of general taxonomicl chemical and molecular data sources in mostly un/semi-structured formats. Informatics strategies to extract, integrate and analyze such heterogeneous information may be important to discover implicit relations and formulate hypotheses pertaining to possible medicinal applications ofplants.
This dissertation aims to apply informatics approaches to identify potential plantbased leads of pharmacological importance. The hypothesis is that informatics approaches can be used to integrate available information from disparate sources containing plant knowledge and can thereby lead to the prioritization of information associated with plants for the identification of potential drug-discovery leads with pharmaceutical importance. This work starts with a survey of bioinformatics applications for identification and study of medicinal plants, which highlights opportunities and challenges in developing time or cost effective leads. Information extraction and semantic integration techniques were then adapted for plant-specific information to discover meaningful information from data.
This work brought together information from within a number of electronically accessible heterogeneous data sources from ethnobotany (e.g., Dr. Duke's phytochemical and ethnobotanical databases) and biomedicine (e.g., MEDLINE, PubChem, KEGG, ClinicaITrials.gov). A concept-based analytic approach was then developed to reveal direct (i.e., plant concepts directly correlated with disease concepts) and inferred (ie., plant concepts associated with disease concepts based on shared signs and symptoms) relationships from within literature knowledge sources like MEDLINE. Finally, a feasibility study was done that demonstrated the potential to bridge ethnobotanical data with contemporary experimental data (e.g., biological assays indexed in PubChem), focusing on an evaluation of the medicinal use of plants identified from Micronesian islands (Palau and Pohnpei).
From the studies conducted in this dissertation, several interesting phyto-therapeutic leads were identified that may serve as candidates for future investigation. The results from this dissertation thus highlight the applicability of informatics methodologies for bridging ethnobotanical and biomedical knowledge to identify and prioritize plants or their phytochemicals of potential therapeutic importance.
This dissertation aims to apply informatics approaches to identify potential plantbased leads of pharmacological importance. The hypothesis is that informatics approaches can be used to integrate available information from disparate sources containing plant knowledge and can thereby lead to the prioritization of information associated with plants for the identification of potential drug-discovery leads with pharmaceutical importance. This work starts with a survey of bioinformatics applications for identification and study of medicinal plants, which highlights opportunities and challenges in developing time or cost effective leads. Information extraction and semantic integration techniques were then adapted for plant-specific information to discover meaningful information from data.
This work brought together information from within a number of electronically accessible heterogeneous data sources from ethnobotany (e.g., Dr. Duke's phytochemical and ethnobotanical databases) and biomedicine (e.g., MEDLINE, PubChem, KEGG, ClinicaITrials.gov). A concept-based analytic approach was then developed to reveal direct (i.e., plant concepts directly correlated with disease concepts) and inferred (ie., plant concepts associated with disease concepts based on shared signs and symptoms) relationships from within literature knowledge sources like MEDLINE. Finally, a feasibility study was done that demonstrated the potential to bridge ethnobotanical data with contemporary experimental data (e.g., biological assays indexed in PubChem), focusing on an evaluation of the medicinal use of plants identified from Micronesian islands (Palau and Pohnpei).
From the studies conducted in this dissertation, several interesting phyto-therapeutic leads were identified that may serve as candidates for future investigation. The results from this dissertation thus highlight the applicability of informatics methodologies for bridging ethnobotanical and biomedical knowledge to identify and prioritize plants or their phytochemicals of potential therapeutic importance.