Get the gene name that related to your interested disease from PubMed Abstract.
GoFrame is an application development framework of Golang.
html implements an HTML5-compliant tokenizer and parser.
htmlquery supports HTML document query.
XPath is Go package provides selecting nodes from HTML or other documents using XPath expression.
regexp2 is a regex engine in pure Go based on the .NET engine
csv is a package related to read and write csv file\
Type in disease name and abstract number, and you will get disease related gene name and other paper information in a csv file. Default disease name is Alzheimer's and default abstract number is 10.
$ go build
$ ./GeneDetectorOR
$ go build
$ ./GeneDetector -disease diabetes -n 20A CSV file with the related gene symbol. There are other information in this csv file, including: paper title, url, abstract content, gene name, pmid, doi, keyword(disease name)
The main strategy to get the related gene name is use regular expression to match the gene symbol. I didn't use text mining strategies to get the gene name due to my limit knowledge background. But I do spend time to get to know more about the knowledge of text mining. I found that Named-entity recognition(NER) would be a good strategy for the next step of this project.
Thanks for Robin L. having good discussions about the tips of web scrap!
Thanks for Professor Carl Kingsford importing all the great knowledge about golang!
Thanks for TA Siddharth Reed for giving me lots of help when I found this project was really hard to go on!
Thanks for TA Jingjing Tang for grading and giving feedback!