Example for use of checkboxes in PubCrawler results file

This example tries to explain the usage of checkboxes in the PubCrawler results file and its limitations.

Imagine your search profile consists of the following four queries:

pubmed 'Watson' ( SCIENCE [Journal Name] OR Nature [Journal Name] ) AND Watson JD [Author Name] BUTNOT Crick F [Author Name]
pm_neighbour 'PubCrawler Paper' 10529811
genbank 'long sequences' ( human [Organism] OR mouse [Organism] ) AND 200000:350000 [Sequence Length]
gb_neighbour 'SSK1' 5915906
The first one looks up papers in Science and Nature that have been published by JD Watson but not by F Crick. The second one looks for articles related to the paper with the PubMed ID 10529811 (Paper on PubCrawler). The third one looks for sequences from human or mouse of length 200000 to 350000. The fourth one looks for sequences in GenBank which are related to entry GI5915906 (SSK1 gene).
This means that queries 1 and 2 search in PubMed whereas queries 3 and 4 search in GenBank.

Suppose that one day you get new hits for each of the queries. They will show up in your results page as articles and sequences in summary format, like the following:

summary of report

Additionally you will find one or more of the following NCBI button bars in the results file:
NCBI button bar

Each of the buttons will by default apply to all of the reports from the first database (here PubMed) that are shown in summary format. Selecting 'MEDLINE' instead of 'Summary' and clicking on the 'Display' button will bring up all the PubMed articles in the specified format, as if all their checkboxes had been selected.
If you are only interested in a certain subset, then go back, click on the checkboxes of the desired reports (selections from all new PubMed hits are possible) and click the 'Display' button again.

The selector of the 'Show:' field is automatically set to the number of unique new hits in summary format from the first database (duplicate entries sorted out).

Note: New hits which exceeded the fullmax value are not selected by the action described above! To work on them they have to be accessed through their links (MORE: 0-day-old records).

If you select hits from the section of the second database (here GenBank) and click one of the NCBI buttons, their identifier number will be mapped to the wrong database and you won't get the desired reports.

Why is that? NCBI uses hidden variables to store the UIs of documents on your browser page and the database where they come from. Normally each section in the PubCrawler results file, which contains NCBI output, would overwrite the hidden variables of the previous one, which means that the buttons would only work for the last section with full reports. We have modified the output so that the buttons will apply to all reports from the first database. Application to both databases is not possible at the moment and the solution we are providing should satisfy most user's needs.

However, if you get results from more than one database there is a way to efficiently process results of the second one. In our example, the sequences from GenBank that appear in summary format can be retrieved all at once through a link at the bottom of the results file:

NCBI button bar

Clicking on the blue, underlined link will open a new window which brings up all the GenBank hits in a pure NCBI environment (no PubCrawler text). This will allow you to process them further, i.e. choose different formats, add to clipboard, etc.

If, at another day, you get only new hits for the searches at GenBank (queries 3 and 4), then GenBank is the first (and only) database from which reports in summary format are shown and checkboxes and NCBI buttons work as described above.


This is not a perfect solution but the best we can do at the moment and we hope that you are happy with it!





[PubCrawler Home Page]     [PubCrawler Program]     [PubCrawler WWW-Service]     [PubCrawler Configurator]



Last modified at $Date: 2000/02/28 22:33:25 $
For questions or remarks, please send mail to PubCrawler.