textsearch searches for words (specified as a regular expression) in the description text of one or more input sequences. It writes an output file with optional contents such as the name, description and accession number of any sequence whose description line from the annotation matches the search term. Optionally, the search is case-sensitive and the results output as an HTML table. textsearch is convenient for small input files but will be slow for larger files and databases; you should use use SRS or Entrez instead.
textsearch searches only the description line, not the full sequence annotation.
|
|
The first column in the name or ID of each sequence. The remaining text is the description line of the sequence.
When the -html qualifier is specified, then the output will be wrapped in HTML tags, ready for inclusion in a Web page. Note that tags such as <HTML>, <BODY>, </BODY> and </HTML> are not output by this program as the table of databases is expected to form only part of the contents of a web page - the rest of the web page must be supplier by the user.
The lines of out information are guaranteed not to have trailing white-space at the end. So if '-nodesc' is used, there will not be any whitespace after the ID name.
This is a rather slow way to search for text in databases. If you are searching for text in public databases, you should consider using either Entrez (http://www.ncbi.nlm.nih.gov/Entrez/) or SRS (http://srs.rfcgr.mrc.ac.uk/ or http://www.sanger.ac.uk/srs6/ etc.) instead.