First you can look for the entry name of the proteins in a database such as UniProtKB (http://www.expasy.org/proteomics), and then with these protein identification codes for mouse and human you can use an alignment tool (i. e. SIM Alignment Tool), in order to compare both protein sequences. The comparison will show you how conserved both proteins are.
Search for your protein at http://www.ensembl.org/index.html
Once you select the gene of interest from the results page, then you will see 'orthologues' on the left panel. If you click through that link you can see what other species this gene is present in.
Find the species you are interested in and then look at the column titled 'Compare'. You will see links for alignments (both protein and DNA).
Hi Sue,
Frankly, if this were my project I would begin by downloading both mouse and human coding DNA genome references from NCBI, and handle the problem from the command line. This solution assumes a list of proteins of interest that is small enough to handle manually - say less than 50.
Once I had the two genomes, I would prepare a reference file that includes the nucleotide sequences of the proteins in question, in .fasta format. Next use blast to build a data base off of your protein file. Once this was set up, I would query each genome against the protein reference, separately. Check each output and select the features that align really well. If you get good alignments for both mouse and human to the same protein, then you have some basis to argue that this is a conserved protein structure.
The advantage of this method is that you will have output that gives you % identity and shows you where you might have a snip. Also you will not be depending on how well maintained some online database is.
If the protein is derived from a house-keeping gene, it is mostly conserved. Otherwise, you can blast the protein against human/mouse protein database (online uniprot or genbank), get all closely related homologs and then align them using clustal.
Dear Sue,
The find protein conserved domain (CDD) is possible to perform this analysis, too, in:
- (http://www.expasy.org/proteomics
>> InterProScan • Family and domain datatabase search
http://www.ebi.ac.uk/Tools/pfa/iprscan5/
Input all sequences....
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics
- Bioinformatic Tools