RDoC Expert Survey Author Finder

Chris Iyer Updated 3/23/2023

This code uses pubget and Entrez from BioPython to search PubMed Central (PMC) for articles pertaining to a given keyword/task. In order to use this code, you'll have to pip install pubget and pip install Bio.

What this code does: These articles are downloaded locally and correspondence emails are extracted. Then, Entrez searches for the number of other PMC papers that cite a given paper, and we select the top <n> (100?) papers. The correspondence emails of these papers are then written to a txt file and returned.

Files:

• author_finder_functions.py contains all the functions performing the steps above. Critically, it also contains a dictionary task_keywords that decides what to query PMC with based on which task we are searching for. Edit this dictionary to refine results.
• author_finder.ipynb is where you will put in a ROOT_PATH to write outputs to, and a task_to_run to search. You may either call each function from author_finder_functions in sequence or use run_author_finder() to get the emails in one step. Doing so for each task takes somewhere around 5 minutes.

Current problems:

1. The keywords are imperfect. Including "task" will exclude a lot of good papers, but leaving it out means we get articles like this article that are about the 'stop codon acting as a stop signal' and not about the stop signal task at all.
2. A few emails are lost (somewhere in the ballpark of 2%). I'm not worried.
3. We are sorting by # of citations only of other papers in PMC (not necessarily all citations of the paper, but just the ones in PMC).
4. The search results are not lining up to what I get when I manually search using the same query in PMC Advanced Search. I'm looking into this.