Several researchers and clinicians use multiple databases and search tools to retrieve the required information. Such resources include Google, Google Scholar, PubMed/MEDLINE, Cochrane Central Register of Controlled Trials (CENTRAL), EMBASE, Allied and Complementary Medicine (AMED), Scirus, HighWire Press, BIOSIS, CancerLit, Cumulative Index to Nursing and Allied Health (CINAHL), CAMbase, the MetaRegister and the National Research Register. Google Scholar, Scirus and HighWire Press are of particular importance for scientific literature search.

Though PubMed is the most preferred scientific literature search engine, it is at a disadvantage since it does not scan the entire document. Full-text scanners such as Google Scholar, Scirus and HighWire Press possess this feature and hence, may provide several advantages over PubMed. This ability of the full-text scanners were assessed by an additional comparative study. The study involved a search for specific information regarding the methods employed in various/ specific studies.

Two searches were carried out in this aspect with the following specific objectives:

1. Protocols of RNA isolation from the human testis for microarray experiments.
2. A search for most often used primer designing tools/software.

In the process, a highly customized query for each of the objective was designed, making the best use of available features of each of these tools. The results were obtained by analyzing the first 100 hits from each of the considered tools except PubMed, where the hits analyzed was a combination of various complex queries. In case of PubMed, a comprehensive query was designed using several synonyms of the query components. Each of the components were then tried by limiting the searches to different search fields available in the tool like ‘Title’, ‘Title/Abstract’ etc. These searches were later combined using the ‘history’ option to get a single list of citations. The citations that were common to different searches were obtained, and were analyzed (see Supplementary notes 9).

The study involving the first objective was further refined by comparing the competency of the above mentioned tools in retrieving RNA isolation protocols, involving trizol (specifically), for microarray work.

Note: Brij.in was not considered in the study, since it truncates the results without any indication of the total number of hits. Furthermore, it was found that the citations retrieved by the tool were most often covered by Google Scholar, though the initial trend seemed different. PubMed Central was considered for one of the initial searches regarding primer designing program/software, but was found to be ineffective.

copyright © shodhaka life sciences private limited; all rights reserved