logust79 / pubmedBatch

Launch pubmed quries in batch

pubmedBatch

What it does

• Search gene terms associated with phenotypes in batch on Pubmed. The gene terms shall be grouped in a column in a csv file.
• In order to speed up the query, and to reduce the Pubmed query frequency, the app saves each unique searches in its SQLite database for 14 days (this can be changed. See Installation).

Installation

• To install, run install_dependent.sh first (only work on Linux-based systems)
• If you run into make problem, you might need to install make first, by sudo apt-get install build-essential.
• To change the life time of a query search, go to config.yml and change the value of life in seconds.
• For multiple workers, please use starman: starman --workers 10 -p 3000 -a bin/app.psgi.
• Port default is 5000. You can change it in bin/app.psgi, set port => 8080. Not relevant if you use starman.
• After running, you can open your browser and visit the page by localhost:5000/batch_pubmed.
• Note that relevant data will be saved / fetched according to the username. The data are saved in batch_pubmed_result/username.
• The prefilled fields (AND and OR) can be configured in config.yml.
• You can use the files in testfiles to play with it.

NOTE Please follow NCBI policies. Try to avoid massive queries during States's busy times. Early mornings and weekends are good times for massive queries.

Caveat

• Searched result is automated saved under the username's folder.
• del row will delete row on the displayed data, but will not delete it from your file. So reloading the file will revive the deleted row. I can make it to delete the data permanently though.
• You need to refresh the page to find any recently saved data in the left column.
• Pubmed doesn't want to be harassed too frequently in busy times, so try to use it for massive queries in US night times and weekends.

Disclaimer

This work contributes to the UK Inherited Retinal Dystrophy Consortium.