Bi-directional Methodology for Literature Extraction from PubMed Abstracts using Web Scrapper and Web Crawler

Searching of literature is the prime processes in any research to gain insights on the progress carried out on specific problems. Literature search in biological science and bioinformatics is more complex as the researchers has to find the relevant articles specified to the biological components like DNA, RNA, Gene, PROTEIN. Pubmed is a literature database which consists of millions of articles submitted by researcher based on the experimental studies. Analysing Gene for the study is important for studies and experiments related to diseases. In this paper a bi-directional methodology is purposed to extract literature from pubmed abstracts using web scrapper and web crawler by providing gene names and map gene with Gene card Database. In the second approach a methodology is provided to analyse Gene expression profiles from GEO database using EDA and Map the significant gene retrieved to the Pubmed and Gene Card database. The differentially expressed genes are grouped after performing PCA and clustered using k-means. Lung cancer is major cancer reported in WHO and hence in this study the Lung cancer is used as the use case to test the methodology proposed. The experiments are implemented using Python. From the experimental results it is found that the purposed approach reduce the time for searching the literature.