BLASTN setting help?

[SDPA_Pet]

Hi, I am new to NCBI local blast. I want to BLASTn against nt database. I want to set this criteria.

1> Alignments with at least 200 positions
2> at least 98% identity to subject sequences
3>bit score at least 50

Anyone can type the command for me here, please?

Thanks,

Pet

[skycreative]

I know it is quite easy to filter the result under your criteria. Maybe you can process the result after full run

[SDPA_Pet]

Can you write a command line example for me?

Thanks,

[rhinoceros]

http://www.ncbi.nlm.nih.gov/books/NBK1763/#_CmdLineAppsManual_Appendix_C_Options_for_

[SDPA_Pet]

http://www.ncbi.nlm.nih.gov/books/NB...C_Options_for_

Hi, if I only consider alignments with at least 200 positions, should I use "-num_alignments 200". If I set "-num_alignments 200", the result will give me the first 200 sequences close to my query sequences. OR BLASTN only blast my query sequence longer than 200 bp.

I am not sure what is the function of "-num_alignments". Can any one explain it? I want to set "-max_target_seqs". However, I can use both "max_target_seqs" and "-num_alignments" in my blastn command line.

[GenoMax]

Hi, if I only consider alignments with at least 200 positions, should I use "-num_alignments 200". If I set "-num_alignments 200", the result will give me the first 200 sequences close to my query sequences. OR BLASTN only blast my query sequence longer than 200 bp.

I am not sure what is the function of "-num_alignments". Can any one explain it? I want to set "-max_target_seqs". However, I can use both "max_target_seqs" and "-num_alignments" in my blastn command line.

"-num_alignments n" is going to report alignments for first "n" sequences that satisfy other search criteria you have set, not "n" positions in individual sequences

As stated in the blast command manual: "max_target_seqs" is not compatible with num_descriptions or num_alignments.

[SDPA_Pet]

"-num_alignments n" is going to report alignments for first "n" sequences that satisfy other search criteria you have set, not "n" positions in individual sequences

As stated in the blast command manual: "max_target_seqs" is not compatible with num_descriptions or num_alignments.

Thanks, do you know any settings can control the "n" positions in individual sequences (the length of sequences that I want to blast).

[GenoMax]

Thanks, do you know any settings can control the "n" positions in individual sequences (the length of sequences that I want to blast).

Can you clarify that? You want the target sequence to be "n" nucleotides or longer?

[SDPA_Pet]

I have a fasta file which has unaligned metagenomic contigs. For example, the length of contigs (sequences) in the file may vary from 50bp to 1000bp. Let's say, I would like to blast all the contigs that long than 100bp against nt database. I don't care about the sequences shorter than 100bp. Is it possible to do it.

[GenoMax]

If you have not done the blast yet then you can filter your contigs file to exclude the small sequences by using a script such as: http://www.bioinformatics-made-simpl...ir-length.html or this http://bioinfofly.wordpress.com/2012.../binawk-f-nex/

If you have already completed the blast search then you should be able to parse the output file so that you select entries where the query would be longer than 100. e.g. http://www.bios.niu.edu/johns/bioinf...ast_parser.htm

Note: If your query file has fasta sequences that wrap around then look at this thread before using the scripts to filter the reads by size/length.