Setting up blastn to retrieve FULL query sequence
1
0
Entering edit mode
3.2 years ago

How can i specify standalone blastn options for it to retrieve full query sequences in output?

For example i have a query lengths ~(100 - 300) bp, and i'm blasting one of them (i am doing 2 by 2 blast) against subject with length ~3000bp. But in the output i have a lot of hits that are very-very small, and i'm not interested in that

ex_chr2L_10459384_10459517_+_ cl_chr2L_10459280_10459880_+_2_s_apiMel3_0_659_+_659 88.235 17 2 0 69 85 512 528 0.029 21.4 ACAATGGGCGGTGATTT ACAATGGGAGGAGATTT ex_chr2L_10459384_10459517_+_ cl_chr2L_10459280_10459880_+_2_s_apiMel3_0_659_+_659 85.714 14 2 0 100 113 294 307 1.4 15.9 CTCAAGTTTACTAT CTCTAGTTTATTAT ex_chr2L_10459384_10459517_+_ cl_chr2L_10459280_10459880_+_2_s_apiMel3_0_659_+_659 79.167 24 3 2 50 73 328 349 1.4 15.9 CACCAACAATATCATTGTCACAAT CATCAA-AAT-TTATTCTCACAAT ex_chr2L_10459384_10459517_+_ cl_chr2L_10459280_10459880_+_2_s_apiMel3_0_659_+_659 100.000 7 0 0 109 115 31 37 4.9 14.0 ACTATAA ACTATAA ex_chr2L_10459384_10459517_+_ cl_chr2L_10459280_10459880_+_2_s_apiMel3_0_659_+_659 100.000 7 0 0 104 110 88 82 4.9 14.0 AGTTTAC AGTTTAC ex_chr2L_10459384_10459517_+_ cl_chr2L_10459280_10459880_+_2_s_apiMel3_0_659_+_659 100.000 7 0 0 64 70 151 157 4.9 14.0 TTGTCAC TTGTCAC ex_chr2L_10459384_10459517_+_ cl_chr2L_10459280_10459880_+_2_s_apiMel3_0_659_+_659 100.000 7 0 0 80 86 183 177 4.9 14.0 TGATTTT TGATTTT ex_chr2L_10459384_10459517_+_ cl_chr2L_10459280_10459880_+_2_s_apiMel3_0_659_+_659 100.000 7 0 0 107 113 268 262 4.9 14.0 TTACTAT

here is the command line options i use:

blastn -query query.fa -subject subject.fa -word_size=4 -outfmt="6 qseqid sseqid pident length mismatch gapopen qstart qend sstart send evalue bitscore qseq sseq" -out output.txt

blastn alignment • 1.3k views
0
Entering edit mode
1
Entering edit mode
3.2 years ago

blastn is a local alignment algorithm so if you only care about full length alignment you need to look for a global alignment algorithm (e.g. Needleman–Wunsch as in the EMBOSS needle program or exonerate with the affine:global option). You could play around with some blastn parameters to approximate the same thing but this depends on defining more precisely what you want and don't want. The alternative, often more straightforward, may be simply to post-process the output to extract what you want.