found strange BLAST behavior... can anyone explain?

[bioinformatics_ftw]

(first post here -- please be kind)
Hello all!!!

I've just come across something strange that I am unable to explain. If anyone could enlighten me it would be greatly appreciated. It is related to the translation step in using blastx, tblastn, or tblastx.

I have a single nucleotide sequence which I have turned into a nucleotide database. If I try to align this sequence to it's own database, I would expect to get a perfect alignment. When I do this using blastN, I do get a perfect alignment. However, when I try to align this sequence to it's own database using TblastX (translated nucleotide sequence to translated nucleotide database) I do NOT get a perfect alignment, I get only a partial alignment. What is it about the translation step that causes this?

Does anyone know why this would be? I'm at a complete loss...

Thank you very much for your expertise &/or ideas!!


Here is the sequence:

ATGAAGAACTTCCTCATCCTTGCCCTCCTTGCCATGGCGGCGACCATGGCCACTGCCCAGTTTTACCCTAGCGAACAATATCAGCCATATCCTGAGCAACAACAACCATTTCTACAACAACAGCCATTGTTGCAGCAACAACAATTGTTGCAGGTTCTGCAGCAACAGCTGAACCCATGCAGGCAATTCCTCGTGCAACAGTGCAGCCCAGTGGCAGAGGTGCCGTTCCTCCGGTCGCAGATCCTGCAACAGAGCAGCTGCCAGGTGATGAAGCAACAGTGCTGCCGGCAGCTGGCACAGATCCCCGAGCAGGTTCGGTGCCCGGCCATCCATAGTGTCGTGCAAGCCATCATCTTGCAGAAGCTGCAACAACAGTTGCTCCAGCCTCAGCTGCAACAACAGCTGCTCCAGCCTCAGCTGCAACAACAGATTCTCCAGGCTCAGCTCCAACAACAGCTGCTCCAGGCTCAGGTGCAACAACAGCTGCTCCAGCCTCAGGTGCAACAACAGCTGCAACAACAGTTGATCCAGCCTCAGCTGCAACAGGTTTTCATCCCGCCTCAGCTGCAACAGGTTTTCCAGCCTCAGCTGCAACAGGTTTTCCAGCCTCAGCAGCAAGCTCAGTTCGAGGGGATGAGGGCGTTTGCACTGCAGGCCCTGCCGGCGATGTGTGATGTGTATGTTCCACCGCACTGCTCCGTAGCCACCACCCCGCTCGGTGTGGTC