I used blastp and DELTA-BLAST algorithms for the same protein sequence, but I am confused with the results. Putative conserved domains have been detected by blastp, but nothing has been detected by the DELTA-BLAST algorithm. I know they used different algorithms. However, can we say there is a putative conserved domain in this protein sequence or not?
Acutally NCBI provides a specific service to search for conserved domains. Here is a link: http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi
You should note that there are several different domain databases to search against. There is the CDD (which is used by Delta Blast to enhance its homology search), but there are also PFAM, SMART and several others. While the CDD description states that it imports data from other databases (including SMART and PFAM), I am not 100% sure that it contains the whole PFAM and SMART database, and it might be worth to double-check by running your protein sequence against different databases.
I think that you should decide whether the protein has a conserved domain or not based on the score for the results that do find the domain. A domain may not be found for many reasons, for example, if there is no such domain in the specific database you search against, but it could be present in another.
If NCBI CDD hit data indicates that the domain represents the protein of your interest but Pfam does not support the same results for the same protein. In such situation which one is more reliable? Pfam or NCBI CDD?
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics