Homework 5 (aka Project)

Due: Thursday, December 12, 2024 at 11:59 p.m.
Points: 100

Homework 4 had you produce a list of publication IDs from a keyword search on PubMed. The final project is to produce a list of the publication citations for that keyword.

Begin with your program 4. First, modify it so that after it requests the keywords, it asks the user how many references to print. You will have to read in an integer for this. Be sure the integer is a positive number!

The program will print out a query string that is a URL of a list of PubMed publication IDs. Use that URL to get the metadata. The web page you get back is an XML document giving details of the publications.

Your job is to print a bibliography from this record. Your entry for each journal should look like this:

A. Bester, R. Zelazny, and H. Ellison, “On the Role of Viruses in Future Epidemics,” Journal of Irreproducible Results 3(4) pp. 29–35 (Mar. 2103). DOI 12.1119/2847595; PUBMED: 23456789.

Then print the abstract, if it is present in the record.

If there is no DOI, omit that part of the entry.

You will need to look at the XML records to get the fields. These are delimited by tags with attributes, each of which may have a value. For example, the element

<ELocationID EIdType="doi" ValidYN="Y">10.1016/j.vaccine.2015.04.071</ELocationID>

has a tag of ELocationID, attributes of EIdType (with value doi) and ValidYN (with a value of Y), and the field contains 10.1016/j.vaccine.2015.04.071, which (as the EIdType value indicates) is a DOI.

The fields of interest will have these tags:

Article — contains the Journal, ArticleTitle (article title), Pagination (page numbers), ElocationID, which gives both the DOI and PII (if those exist), the Abstract, and the AuthorList.
Journal — this consists of several elements, including JournalIssue, which contains the Volume, Issue, and PubDate (publication date), and Title (article title).
AuthorList — this lists the authors, each author being in a field called Author. Subfields of interest are LastName and Initial (the initial of the first name)

Those will be enough to build the reference, as described above.

The easiest way to see what an entry looks like is to ask for it. To get the entry for item with id n, use the following URL:

http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=pubmed&retmode=xml&id=n

It will be in XML, and you can easily see the structure.

You can find methods for processing XML in the Python Library Reference at https://docs.python.org/3.7/library/xml.etree.elementtree.html

Matt Bishop
Office: 2209 Watershed Sciences
Phone: +1 (530) 752-8060
Email: mabishop@ucdavis.edu

ECS 235A, Computer and Information Security
Version of November 26, 2024 at 1:09PM

You can also obtain a PDF version of this.