I would like to search for a 15nt string in which the last 2nt are absolutely matched against the genome of interest, yet I can tolerate 1-2 mismatches in the intial 12nt and any nucleotide in position 13. Later I would like to complicate things by adding a further string of 8nt, preceding the initial string which, although there is no requirement for a match in these nt, I would like to see where any matches do exist?
Initially I searched for a string "XXXXXXXXXXXXNXX" in which the X's represent my sequence and the N, any nucleotide. However the results seemed to ignore the last 3 nucleotides where, upon further analysis of one match, I saw there was a perfect match across all 15nt. Even if this had worked perfectly however, it would not have taken into account my preferences for these last 2nt to be absolutely conserved and only a 1-2nt mismatch in the first 12nt.
I have been using NCBI BLAST.
Any pointers would be greatly appreciated.
I would recommend the UCSC genome browser at, http://www.genome.ucsc.edu/
Then use the BLAT function, it is much better and faster at finding short oligonucleotide matches in the entire genome.
Thank you for the advice Gregory, unfortunately it too struggles here. I generally quite like BLAT as a tool but I run into a couple of problems. First is that BLAT is limited to searches of 25 bases or more and second, that the genome I'm looking at is out of date on the UCSC genome browser (and won't come annotated with gene data). I'm looking at the Anopheles gambiae genome, I would usually use VectorBase/NCBI Blast, but it's not performing properly. If I search for the sequence GCATCGAGCTGANGG for example, there is a perfect match in the gene AGAP009290, yet it will only find this match if I were to replace the N with an A and the match is found in the transcript database but not the genome (despite being part of a single exon, existing in the genome as well as the cDNA).
- Badges
- Science topic
- Similar topics
- Biological Science
- Genetics