Plowing New Ground: Finding Patent Information in Agriculture and Life Science Databases

Brian Carpenter, Patent and Trademark Librarian
Texas A & M University
[email protected]


On January 11-15, 2000, Texas A&M University hosted “Continuing the Legacy: New Horizons,” the 2000 Agriculture Program Conference. At this conference, the Evans Library Science and Engineering Services librarians presented an orientation to electronic databases related to agriculture and life sciences that are available through the library’s Public Access Menu (PAM). In an overview of these databases, attendees were shown sample searches and search strategies that would help them locate patented inventions and perform their research. IIn this article, we will provide: (1) Tips on how to best use some of these databases to find patent information related to agriculture and life science; and (2) Tips on how to search the U.S. Patent and Trademark Office (USPTO) Web-based databases.


Patent Information; Science Databases; Patent Databases; Patent Searching; Search Strategies; Keyword searching; Patent Classification Searching.

Statement of the Problem

Patent searching has always been viewed by academia as either a necessary evil or a truly almost mystical process used to determine if an invention is patentable. The purpose of our research is to show that one does not have to be skilled in the black arts or possess a legal degree to do a fairly competent preliminary patent search. By starting with the tools a professor might be most familiar with, namely commercial databases in his/her own field of expertise, anyone can use search results from these scientific databases as starting points to a preliminary patent search.

Do scientific commercial databases provide the kind of information that can help a researcher begin a preliminary patent search? If this is the case, what kind of search strategy works best?

Goals & Objectives

Our goals are to show how to:

  • Become more comfortable with patent searching.
  • Find patent information in agricultural databases.
  • Supplement information from traditional agriculture literature with patent information.

To achieve these goals, our objectives are to show how to:

  • Determine agriculture databases to search for patent information.
  • Construct effective searches.
  • Spot useful numerical patent information.
  • Broaden the search for relevant records.
  • Use numerical patent patent information to locate patents in USPTO databases.
  • Use appropriate field codes to search USPTO databases.

Literature Review

The idea of using commercial databases for preliminary patent searching began in the late 1980s with the use of online databases through DIALOG and BRS. Snow (1989, 41) first talked about using these commercial resources and their subject databases for patent research. Despite the promise shown by these resources, little was written about the value of using these resources to help with patent searching.

In the late 1990s, the widespread availability of commercial subject databases on compact disc, read-only memory (CD-ROM) and mounting of these databases on local-area networks (LANs) led to a renewed interest in subject-based patent searching. Carpenter, Hart, and Miller (1998, 22) pointed out that a “field-specific [search] strategy” could help researchers “jump start” their preliminary patent searches. Using this strategy, in addition to the seven-step patent search process recommended for inventors by the United States Patent and Trademark Office (USPTO), has provided another way for researchers to do a more thorough preliminary patent search.

The “field-specific search strategy” evolved into pinpointing field codes/access points of commercial subject databases (e.g., notes field – NT) that can help a researcher find the appropriate patent classification. Hart, Carpenter, and Miller (1999) note that increasing numbers of licensed, commercial subject databases are including patent information. Still, the most useful combination of field codes/access points and specific keywords remains unclear. As a rule, researchers tend to like the commercial databases because they allow them to use keywords and Boolean logic to find information. The USPTO has responded to this tendency with their Web-based patent-research databases. Crawford (1999) illustrates that keyword searching alone will not guarantee a successful search in the USPTO databases. For a successful patent search, keyword searching must be combined with a search using the appropriate patent classification information.

Using this information and subject specialists’ knowledge about commercial subject databases available through three primary vendors (Cambridge Scientific Abstracts, OVID, and SilverPlatter), more useful search strategies can be created. The combination of Boolean operators, “subject-specific” keyword searching, and finding the appropriate patent classification can help researchers do more thorough patent research on their topics in a variety of databases.


In our hypothetical scenario, fictitious Professor Bo Vine is a distinguished agricultural scientist at his university. His research requires him to have sufficient amounts of a particular virus that attacks cattle so he can study it in detail. To do so, he uses a technique called viral tissue culture. His new department chair wants faculty members to obtain more patents in the areas of their research.

To obtain a patent, Professor Vine must first determine what patents already exist in the area of viral tissue culture. He must find recent patents and eliminate those for which others have already applied. Unfortunately, Professor Vine has no idea how to search for patent information and is quite apprehensive about doing so. Since he is already somewhat familiar with commercial databases of agriculture information, he decides to search these databases first.

Although he tries very hard, Professor Vine cannot seem to locate any patent information in these databases. Finally, he contacts one of the science librarians at his university and asks for help. The librarian suggests several databases and shows Professor Vine how to search these databases for patent information. Appendix A lists examples of databases to search with some sample search strategies.

Eventually, Professor Vine locates several patent numbers and patent classification numbers in relevant records he retrieved from a database called Agricultural and Environmental Biotechnology abstracts (AEBA). In Full View mode, he notices in the NT (Notes) field of these records that some patent classification numbers appear more frequently than others (e.g., 435, 424, 530). To make his search of this database more complete, the librarian suggests to Vine that he broaden his search for such records. To do so, the Professor goes to AEBA’s Advanced Search and constructs a statement to search the NT field for one or more of these frequent classification numbers. This search retrieves several additional relevant records.

Now, Professor Vine needs to search the USPTO databases for individual patents within these classification numbers. To help him do so, the librarian familiarizes Vine with the basic steps of patent searching. Afterwards, he goes to the USPTO Web site and searches the Classification Index. After searches with “cattle” and “bovine” yield no results, he tries the term “animal” and eventually finds classification 435/235.1. The Classification definition for 235.1 pertains to “media for propagating” viruses and bacteriophages, so this classification appears useful. Professor Vine then searches the Bibliographic and Full-Text USPTO databases for patents related to viral tissue culture. He finds several that he must investigate further to see how much they overlap his area of research.

Results & Discussion

Professor Bo Vine eventually found several patents that pertain to his research into viral tissue culture, especially with regard to viruses that attack cattle. His next step would be to thoroughly investigate these patents to determine how closely they relate to his specific area of research.

The professor had a few minor obstacles to his search for patent information. However, he overcame these with the help of a science librarian at his university. The librarian showed him the following points:

  • The best agricultural databases tp search.
  • How to construct effective search statements in these databases.
  • How to broaden his initial search to retrieve additional records.
  • How to recognize useful numerical patent information within records retrieved from pertinent databases.
  • The basic steps of patent searching.
  • The appropriate USPTO databases to search.
  • How to construct effective search statements in these databases,
    including how to search appropriate field codes.

Professor Vine made an important advance when he discovered the correct term by which to search the USPTO Classification Index: “animal” instead of “cattle” or “bovine.” The important realization is that “animal” subsumes the latter two terms, neither of which have classification that pertains solely to them. Vine now feels more confident and comfortable when he searchesfor patent information.


Professor Bo Vine, despite going down some initial blind alleys (e.g., trying to use the terms “cattle” and “bovine” to locate a proper patent classification), found the patent information that pertained to his research. Using a combination of “subject-specific” keywords and Boolean logic plus a range of patent classifications, researchers can find relevant patents and patent-related documents. The strategy used to find the professor’s information can be used with any commercial subject database to find patent-related information.

There also were some problems identified by this research, which are:

1. Not all commmercial databases have patent information.
2. Some databases that do have patent information do not have patent classification data. The best that can be done with these is to write down patent numbers cited in the particular database and then find the actual patent.
3. Librarians showing commercial inventors and researchers how to use this strategy must be familiar with subject based commercial databases.

Appendix A: Notes, Tips and Strategies

In this Appendix, we provide some additional notes, tips, and strategies on the searches conducted in scientific and patent databases:

I. Search Tips and Strategies

Prof. Bo Vine did his initial search in:

Cambridge Scientific Abstracts’ Agricultural and Environmental Biotechnology Abstracts (AEBA) under the Environmental Sciences & Pollution Management Subfiles

Advanced Search strategy: Use the Notes (NT) field coupled with a keyword (e.g., “cattle”) in the Anywhere field. Remember to set your search range from earliest to the most recent date to ensure that you retrieve all citations that fit your search.

“NT= ((435/*)) and (cattle)” is the way your search will appear when you enter it into the CSA search box.

II. Some databases can be searched by patent number.

These include:

  • BIOSIS [Ovid]
  • Food Science and Technology [CSA]
  • Microbiology Abstracts [CSA]

III. Some databases can be searched by patent number and classification.

These include:

  • Aquatic Sciences and Fisheries [CSA]
  • Biological Sciences [CSA]
  • Environmental Sciences
  • NTIS

IV. The following record field codes can serve as access points for patent information in electronic commercial databases:

AU Author
CS Corporate Source
DT Document Type
ID Identification
NT Note [this is where you will usually find classif. info.]
NU Number
PA Patent Year
PU Publication Type
PY Pubication Year
RD Report Date
RN Report Number
SO Source

V. To help your researcher start to think about how inventions are
classified, have them consider these questions about patents:

What does the invention do?
The “Essential Function” of the invention

What’s the end result?
The “Essential Effect” or basic product resulting from the invention

What is it made of?
The “Physical Structure” of the invention

What is it used for?
The “Intended Use” for the invention

VI. How Professor Bo Vine did his searches in the Patent Bibliographic and Full-text Web-based databases:

Bibliographic database: ccl/435/* and (cattle or
bovine) and “tissue culture” (Make sure to specify that you want to search 1976-2002).

Full-text database: clas/435/$ and (cattle or bovine) and “tissue culture”
(Professor Vine only searched 1999-2000 patents. If he had searched 1976-2000, he would have found a lot more. Also remember that truncation of search terms can be helpful. Vine did not use it in this search, but he might try it in later research now that he is feeling more comfortable with how to do preliminary patent research.)


Snow, Bonnie. “Patents in Non-Patent Databases: Bioscience Specialty Files”Database, 12 No. 5, (October 1989): 41-48.

Snow, Bonnie. “Patents in Non-Patent Databases: Food, Agriculture and Environment Files.” Database, 12 No. 6, (December 1989): 115-119.

Carpenter, Brian B., Hart, Judith L., and Iller, Jeannie, “Jump Starting the Patent Search Process by Using Subject Oriented Databases.” Database, 21, (December 1998): 20-23.

Hart, Judith L., Carpenter, Brian, and Miller, Jeannie, “The Patent Search Process Using Subject-Oriented Databases,” Encyclopedia of Library and Information Science (Forthcoming)

Crawford, Julia. “Keyword Searching on the USPTO Web Databases,” In The Farm System-PTDL Joint Panel (4th Annual Independent Inventors’ Conference) Houston, Texas. September 24, 1999.