blastn not acknowledging "N" string.

Question

[robbenstijn]

My sequence of CCNCCNTNNCCNC does not show any results when I use it as input in blastn on the human genome, although I am a 100% sure that results should come up. Changing the "N" string with any other of the nucleotide strings does give valid results, so I suspect that the tool just doesn't work using the "N" string. Does anyone know how I can make sure that some results do come up?

[Pierre Lindenbaum]

I don't think blastn is designed for such small degenerate sequence.

For short sequences, word-size must be less than half the query length, otherwise reliable hits can be missed.

Even ,when ignoring the CR/LF, a bunch of hits is found using a simple grep...

$ grep 'CC.CC.T..CC.C' hs37d5_all_chr.fasta | wc -l
117794

[robbenstijn]

Thanks Pierre, I am not really sure what you mean. In the browserversion of BLAST, it should show results right?

[cschu181]

If you can use short-blastn instead of megablast in the "browser version" (assuming blast.ncbi.nlm.nih.gov?) then it might work, but as Pierre says, it might be too short and degenerate.