BLAST standalone: length and missmatches

Question

[Jan.V.Henkel]

Hey!

I got a short question:

I used BLAST+ to seach a fasta file containing small sequences (18 bp) against a manual database with 16S sequences using -task blastn-short -outfmt 6

With -outfmt 6 i get a tabular output with length and missmatch (among others).

Question: When i get a length of 18 (that is 100% coverage) and 1 missmatch i would suggest i got 1 missmatch in 18 bp. When i get a length of 17 bp and 0 missmatches, does that mean there is a missmatch at the first or last base?

cheers!

[Jan.V.Henkel]

I made up some sequences with the respective missmatches at the end of the query sequences. Indeed, if the missmatch is at the end the output is "length" 17 instead of 18 but no missmatch is shown.

Do you know whether there is an option to tell blastn to always align the full query sequence?

[Carlo Yague]

By definition, BLAST makes local alignments only (BLAST= Basic Local Alignment Search Tool). To align the full sequence, you will need to use a global alignment tool.

Alternatively, I guess you could try to increase the score penalty for gap opening and gap extension, this will favor full length alignment.

[GenoMax]

Not sure if blat would do better. May be worth a try.

Answer 1

[5heikki]

To answer your original question, column 7, q_start start of alignment in query sequence