Hello everyone,
I have got sequences of the gene I have been working on and its two paralogs. As we know that they are expressed into the plant so I want to know the sequence similarity between two paralogs upto 2 kb 5' UTR. I tried aligning it in several ways, using clustal o, Mega, T-Coffee, MAFFT etc but I am not getting the kind of alignment which I have been looking for, as I get lots of gaps, which are for INDELs, I even downloaded seaviewer program to remove the gaps but in vain, it doesnt work the way I thought it should be. So can anyone suggest me the best way to align paralogs sequences to get maximum information out there? BTW I did pairwise alignment of sequences but when I try to MSA, everything is out of order and then its really tedious to remove the gaps.
Looking forward to some smart way to get it done with maximum evolutionary information.
Thanks.
I'm not sure how all the alignment programs that you mention actually work but I seem to rememfber you can assign penalties for various conditions, one of them being the inclusion of gaps. Try increasing the gap penalty and see what effect that has. Perhaps, as you have so few sequences, just align them manually. From the alignments you already have you know the regions of higher homology. Work outward from these regions and you may find that you are able to align the sequences more easily. Your brain is highly tuned to spotting patterns in a way algorithms cannot.
@Hanley,
What do you mean by aligning them manually? I am not a bioinformatics person so I find it difficult to do.Would be great if you explain a bit more. Thanks!
You say you're looking for a smart way to gain information, unfortunately without a full understanding of the parameters controlling the application of algorithms it's going to be hard to obtain want you want. It's been many years since I routinely aligned sequence (using Phred/Phrap) and in the intervening years I'm sure many more tools have become available in which you might be able to manually align /move/cut/insert/expand sequence (can anyone else offer some suggestions??). Failing that, from the alignments you have already performed you will have been given the coordinates at which each sequence aligns e.g. seq 1, bases 34 - 55, seq 2, bases 83- 104 etc. With these coordinates, and taking one full length sequence as your template, you can then cut and paste sequences into a word document and shift the sequences about yourself (use courier font). This I think helps gives you a 'feel' for the sequence you're looking at and can perhaps reveal patterns not exposed in the alignment programs.
I meant masking for any repeat present in two sequences. First step is to masking repeat regions and than align 1st to see if you see some visible similarities may be promoter regions are conserved. You can break sequences into several blocks of 20-25 bp and align them as suggested above.
- Badges
- Science method
- Similar topics
- Biological Science
- Genetics
- Genetic Analysis
- genetic techniques
- Sequencing