Protocol Online logo
Top : Forum Archives: : Bioinformatics and Biostatistics

alignment problem - (Mar/07/2007 )

Hello everybody!!!
I hope to be as clear and consice as I can...

I had to compare a sequence I've designed on the base of the description included in a publication, with the original one whom one of the authors sent me. I expected them to be identical.
Preliminary, I found that they have exactly the same number of nucleotides (13350). So I proceeded performing an alignment using the tool bl2seq from blast.

When I looked at the output, I found that the original sequence is inverted and complementary to mine.
A part from that, the two sequences match perfectly except a 78 bp-long interruption sited in the middle.
I've extracted that segment in both of them in order to compare it manually.
The result was a perfect alignment, with no gap, sense inversion or wrong complementation...

I can't understand why blast tells me that there is that segment of mismatch between the two sequences, when in fact they are the same.....
Is there any reasonable explanation?
Do I have to compare the blast output with another software result???

I hope you will be so kind to help me understand.....
Thank you very much in advance!!!
ILA

-ila-

QUOTE (ila @ Mar 7 2007, 04:00 AM)
Hello everybody!!!
I hope to be as clear and consice as I can...

I had to compare a sequence I've designed on the base of the description included in a publication, with the original one whom one of the authors sent me. I expected them to be identical.
Preliminary, I found that they have exactly the same number of nucleotides (13350). So I proceeded performing an alignment using the tool bl2seq from blast.

When I looked at the output, I found that the original sequence is inverted and complementary to mine.
A part from that, the two sequences match perfectly except a 78 bp-long interruption sited in the middle.
I've extracted that segment in both of them in order to compare it manually.
The result was a perfect alignment, with no gap, sense inversion or wrong complementation...

I can't understand why blast tells me that there is that segment of mismatch between the two sequences, when in fact they are the same.....
Is there any reasonable explanation?
Do I have to compare the blast output with another software result???

I hope you will be so kind to help me understand.....
Thank you very much in advance!!!
ILA

ILA, my best guess is that the 78 bp region is a so called "low complexity" region, that they are streach of a's or t's or the like. And in you blast2sq settings you set the low-complexity mask on. If you turn the feature off, you might have it 100%.

-cyberpostdoc-

GOOD SHOT!

Thank you biggrin.gif

-ila-