Sciencemadness Discussion Board

Protein sequence

phlogiston - 6-2-2015 at 13:15

For our interest, please share your interest and goal with this.

It is a DNA sequence, not protein.
using nucleotide blast, you get a very good (but not perfect) match with a gene encoding an alpha-dextrin endo-1,6-alpha-glucosidase from the bacterium <i>Raoultella ornithinolytica B6</i>:

Code:
Query CGTTCTATAGATACGCGATGACGGTATACCATCCGCA |||||||| | |||||| ||||||| ||||||||||| Sbjct CGTTCTATCGCTACGCGCTGACGGTTTACCATCCGCA


When you would try to translate your DNA sequence first, you would find that in the frame that you posted, the 7th codon is a stop codon (TGA), so that is probably the wrong frame.

Code:
cgttctatagatacgcgatgacggtataccatccgcaaagt R S I D T R - R Y T I R K


Looking at other frames (include the reverse strand), you will find stop codons in every possible one, except for the reverse complement skipping the first two bases:
Code:
actttgcggatggtataccgtcatcgcgtatctatagaacg F A D G I P S S R I Y R T


pblast does not return a perfect hit for this sequence. The best one is from an inosine-5'-monophosphate dehydrogenase from a bacterium from the <i>Cryocola</i> genus.

Code:
Query FADGIPSSRIY FA+GI SSRIY Sbjct FAEGISSSRIY



Am I just doing your homework for you or is this result in any way interesting?

[Edited on 6-2-2015 by phlogiston]

morsagh - 6-2-2015 at 13:32

I wanted to learn how to use DNA codon table and from sequences match to protein. I am just on second class of high school so truly it isnĀ“t my homework.

mayko - 6-2-2015 at 14:41

If you want to see what peptide a DNA sequence would be translated into, a good web option is the Expasy Translate tool. (Expasy has a lot of other good tools) Enter your sequence, and it returns a codon-by-codon translation, for all three reading frames on the forward and reverse strand. (Keep in mind that only open reading frames, beginning with a Start, ending with a Stop, etc, correspond to 'well formed' peptides.) This would be best if you only have a few sequences to check, occasionally.

If you have more than you can do by hand, my tool of choice is the BioPython module. Seq().translate and SeqUtils.six_frame_translations would do what you want.

If you're looking to search for proteins which are similar to what your sequence would translate to, BLAST is likely the tool for you. In particular, BLASTX will search protein databases, given a nucleotide sequence.

Again depending on your specific situation, BLAST has a web utility and a command line exectutable.

If your sequence comes from a known species with a sequenced genome, you can also use the UCSC Genome Browser's BLAT search to view your sequence in its genomic context.

I do most bioinformatics with BASH and Python scripts; if those aren't your language or they're new to you, check your local library for "Bioinformatics for [Perl/Java/etc]" books.

phlogiston - 7-2-2015 at 14:31

I see, then Maykos post is much more helpful.
I assumed you were trying to identify an existing protein that was encoded by that stretch of DNA.

Although there are many excellent tools online to do the translation for you, it is definitely also a good learning experience to try doing it by hand using a genetic code table

You seem to be interested in biochemistry at an early age then already. Enjoy it, it is a tremendously fascinating topic, and I fully believe the next revolution in science will be in biology.

[Edited on 8-2-2015 by phlogiston]

Mesa - 23-4-2015 at 17:31

Quote: Originally posted by mayko  


I do most bioinformatics with BASH and Python scripts; if those aren't your language or they're new to you, check your local library for "Bioinformatics for [Perl/Java/etc]" books.

My brother started off his postgrad microbiology work using Python etc.
But he had switched over to an old 400mhz UNIX box by the end of the project. Supposedly the obsolete RISC processors only took a fraction of the time to do protein folding calcs etc. than the models released 15 years after them.


phlogiston - 25-4-2015 at 02:23

That link is for a peptide sequencing service, not DNA.
DNA sequencing is indeed not difficult at all, with the right tools.

Chemosynthesis - 25-4-2015 at 05:13

Very true, though those sequencing tools are changing more and more rapidly. Seems like by the time we get a sequencer and trained operator, the next new thing is "in." Glad it's not me!