how to use mafft do find out kir genes?

[arkilis]

recently get a batch of data from miseq, is there way to use mafft to find those kir genes?

I searched online for mafft usage, which looks no where to put kir databases.

http://mafft.cbrc.jp/alignment/software/

many thanks!

[GenoMax]

kir == Killer cell immunoglobulin-like receptors (KIRs) genes? Why not get the sequences you are interested in, use BBSplit to fish out the sequences from MiSeq that map and then do mafft or some other MSA on the sequences of interest.

[arkilis]

Thanks GenoMax.

I tried the whole weekend. BBSplit/BBMap looks like a alignment tool. I have the KIR gene database, need to get the what it has (KIR) in the sequence files. Any clue on this?

Thanks again!

[GenoMax]

Thanks GenoMax.
I have the KIR gene database, need to get the what it has (KIR) in the sequence files. Any clue on this?

Can you elaborate as to what "database" here means? You have KIR genes sequences but in what format (plain text, blast, one of NGS aligner index format)?

BBSplit uses alignments to identify reads that map to sequences of interest and then separate them (this can work the other way around as well i.e. you discard reads that map as being uninteresting and collect unmapped reads to process further).

[arkilis]

Can you elaborate as to what "database" here means? You have KIR genes sequences but in what format (plain text, blast, one of NGS aligner index format)?

BBSplit uses alignments to identify reads that map to sequences of interest and then separate them (this can work the other way around as well i.e. you discard reads that map as being uninteresting and collect unmapped reads to process further).

Sorry for the confusion.

KIR gene from www.ebi.ac.uk/ipd/kir in fasta format. I might want to use this as the database to find out how many of them are in the current sequence files. I read some papers online, which says all the sequences (fastq) need to be assemblied first. http://www.nature.com/nrg/journal/v1...l/nrg3174.html

Newbie to this area. Thanks for your advice!

Ben

[GenoMax]

If your KIR files are in fasta format then here is what you do (command only for reference, make appropriate changes as needed):

If you have single end reads

$ bbsplit.sh in=reads.fq ref=KIR.fa outu=things_that_do_not_match.fq outm=reads_match_KIR.fq

Reads that match things in KIR.fa are collected in "reads_match_KIR.fq" file rest are put in the other file.

If you have paired-end reads:

$ bbsplit.sh in1=reads1.fq in2=reads2.fq ref=KIR.fa outu1=no_match_1.fq outu2=no_match_2.fq outm1=matched_to_KIR1.fq outm2=matched_to_KIR_2.fq

Note: This should definitely work with a single KIR sequence file (it also may work with a multi-fasta file). Give it a try and see what happens.

[arkilis]

If your KIR files are in fasta format then here is what you do (command only for reference, make appropriate changes as needed):

If you have single end reads

$ bbsplit.sh in=reads.fq ref=KIR.fa outu=things_that_do_not_match.fq outm=reads_match_KIR.fq

Reads that match things in KIR.fa are collected in "reads_match_KIR.fq" file rest are put in the other file.

If you have paired-end reads:

$ bbsplit.sh in1=reads1.fq in2=reads2.fq ref=KIR.fa outu1=no_match_1.fq outu2=no_match_2.fq outm1=matched_to_KIR1.fq outm2=matched_to_KIR_2.fq

Note: This should definitely work with a single KIR sequence file (it also may work with a multi-fasta file). Give it a try and see what happens.

Finally I used the command you suggested:

bbsplit.sh in1=read1.fastq in2=read2.fastq ref=KIR_nuc.fasta outu1=no_match_1.fq outu2=no_match_2.fq outm1=matched_to_KIR1.fq outm2=matched_to_KIR2.fq

Will discuss with colleagues and get back Thanks!