Hello everyone, I did the analysis for a whole cDNA, after i did the BLASTX, i found this gene had the high homology with some other genes, then i chose the higher homology of them did the multiple alignments, but the predicted amino acid sequence of this gene just has a little same with others genes, is this normal? what's the problem of that?
You are not giving that much info about the tool that you have use for your multiple aligment neither the lentgh of the sequence that give a high score in blastx. Nevertheless, I will try to give you what I think that might have happened.
When you use blast tools (blastn, blastp or blastx), you are using a local alignment tools. On the contrary multiple alignment tools are performing global alignment. So you can get a high score with blast that reflect a high identity with a region of your protein but have a poor identity score when aligning whole sequences.
Cheers,
SP
Thank u Sandrine Palcy, I used Mega 4 software for multiple alignment analysis fo the whole ORF of those genes. From you idea, i got the results of the multiple alignments is correct, i'm right?
Have a good day!
Alice
The Mega 4 uses ClustalW multiple alignment method, a well-established method and one of the most popular. So if it is applied correctly this method has proven sucessfull.
My idea was that both results from blastx and multipe aligment software are not necessarily inconsistent. If, for exemple, the blastx high score results concerned less than 40% of the total protein length then your multiple alignment will return a low identity score. In brief, you cannot directly compare the results got from local alignment to those got from global alignment.
thanks Sandrine Palcy, so if I want to do the multiple alignment of my gene, what should I do and how can i do that?
I think that you have done one already, isn'it? It's just that you did not understand the result and that what motivated your question here.
May be if you explain what is the beginning and final goal of your investigations, I could perhaps be of help for setting up a strategy...? For now, I have got that you have a gene (you don't have to mention which one!) of interest and your input information is a cDNA sequence to which you would like to find some homology in protein databases for (functional) identification. Right?
Sandrine Palcy, thanks u so much, yes, u r so right, I finished one. I want to clone this gene and get some information of it.
By finding some homology in protein databases for identification of my gene's probable functions, but i can't understand the result of the multiple alignment.
Best wishes!
Alice
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology