Establish a knowledge graph of tumor immunity-related probiotics and automatically predict potential new drugs
Posted in Internship Presentations | Tagged Fall 2021
Mentor: Yongliang Yang (Doctoral Supervisor and Deputy Dean of the School of Life Science and Technology, Dalian University of Technology.)
Date/Time: December 7, 2021 at 8am
Abstract: Knowledge graph is a hot technology in the field of artificial intelligence. To put it simply, the knowledge graph is a visual network that displays knowledge in the form of a graph to represent the interaction between different entities.
The goals of this project were to first, establish a knowledge base of probiotics and diet-therapy which relate to tumor immunity. This can help us find probiotics and diet-therapy that help reduce the side effects of tumor chemotherapy and radiotherapy. Then we used Neo4j software to build a knowledge graph based on this knowledgebase Taking into account various parameters, we have designed a scoring function called “Volia”. This function can make an evaluation of all aspects of data for each product. Then we used Python to write code to connect the knowledge graph with the scoring function. Driven by the knowledge graph, we inferred and designed special health products to reduce the complications of radiotherapy and chemotherapy for cancer patients. In the end, this project introduced a drug that has passed quality testing and is undergoing clinical trials in major hospitals.
The vision of this team is to establish a large-scale medical knowledge graph platform including more than 10 diseases such as tumor immunity, dilation of mammary duct, diabetes, etc., so for each we will need to establish a corresponding knowledge base. Toward this goal, I used the same method I used for the knowledgebase for tumor immunity to build a knowledge base of diabetes and probiotics.