Workflow example: Use BLASTP to search for sequences related to a query protein and retrieve hits

The pea defense gene DRR206 confers strong resistance to fungal pathogens when transformed into Brassica napus. Although no Brassica homologue of this gene has been found by hybridization, we would like to see if a homologous sequence can be found in speices closely-related to B. napus. Since protein searches are more sensitive than DNA searches, we need to get the amino acid sequence from the genomic sequence.

The first step is to read in GenBank file for the genomic sequence. Save the file PSU1716l.gen  in your current working directory. In GDE, choose File --> Open, and read in the file.

Next, we need to extract the protein coding sequence from the genomic sequence. In GenBank files, coding sequences are annotated as CDS features. Choose Database --> Features (extract by feature keys):

Set the feature key to CDS, Database to "Selected sequences", and send output to GDE. A new GDE window pops up with the coding sequence which was extracted from the larger genomic sequence. To translate into protein, select PSU11716:CDS1, and choose DNA/RNA --> Ribosome. Where the menu says "Send output to a new GDE window", choose "No." The translated protein will be sent to the current GDE window.

To run the BLASTP search, select the amino acid sequence and choose Database --> NCBI BLASTP. Choose the GenBank NonRedundant (Protein) database. To limit the search to relatives of Brassica napus, click "Yes" for Restrict search to entries containing string:, and type "Brassicaceae" as the search string. To eliminate poor matches, set the # matches expected by random chance to 1.0e-8 (ie. 10-8). Click on OK to start the search.

The query sequence is sent to the NCBI Blast server, and the results are returned when complete. The results appear in several windows. The Blast report appears in a text editor

and the Accession numbers of the hits are sent to another text editor window:

The Accession numbers are also sent to a dGDE window. dGDE is a program for working with lists of numbers, such as accession numbers, TaxID numbers etc. dGDE is designed to send requests to the SeqHound data warehouse (, and retrieve information from SeqHound. Since SeqHound can only retrieve sequences using GI numbers, we first need to convert the Accession numbers into GI numbers. Choose Convert --> Convert Name/Acc to GI. Click on "Accession to GI", and send the output to the current GDE window (ie. No).

The Accession numbers are sent to SeqHound, and the GI numbers are read into the GDE window. Select the GI numbers, and choose Retrieve --> Protein from Protein GI. Click OK.

GDE will send the GI numbers to SeqHound, calling the SeqHound method SHoundGetFastaList. The sequences are returned to a new GDE window: