I have genomic data (~30000 sequences), including transcript and related CDS sequences. Now, I want to know intron sequences and intron number of each sequences. I want to use Perl/Bioperl program to resolve this problem. Can anybody give me script to run the program?
Why do you want to reinvent the wheel writing perl script to do so.
if you have genes gtf file, convert it to bed format by gtf2bed utility from bedtools then make a genes file in bed (chrom gene_start gene_end from gtf)
substractBed -a genome.bed -b converted_gtf_to_bed.bed > introns.bed
bedtools getfasta [OPTIONS] -fi -bed -fo
hth
Why do you want to reinvent the wheel writing perl script to do so.
if you have genes gtf file, convert it to bed format by gtf2bed utility from bedtools then make a genes file in bed (chrom gene_start gene_end from gtf)
substractBed -a genome.bed -b converted_gtf_to_bed.bed > introns.bed
bedtools getfasta [OPTIONS] -fi -bed -fo
hth
The best approach (perhaps surprisingly) depends on the origin of your data.
If the data comes from an annotated genome, it would be much better to use the annotations as Manvendra suggests to determine intron boundaries and only then download the sequences. However, in that case, you can probably directly download introns or their annotations without any need for substracting with bedtools.
If your transcript data (or both) come from some experiment then you would need to map transcripts to genomic sequences to see where the exon/intron boundaries fall and in effect substract the one set of sequences from the other. You can probably do it with bl2seq if you know which sequences belong together, or something like BLAST or BLAT if you don't. This approach may not determine splice sites precisely. The processing of such output could be done in Perl/Bioperl, but I don't know if there is a ready-made script for exactly this problem.
Why would you care what OS you use to solve it? There are some not-so-easy ways to get BedTools on windows, such as Cygwin, pybedtools or http://www.moosechips.com/2011/02/bedtools-python-install-in-windows/ It is probably much easier to run a virtual linux machine, such as Biolinux under something like VMware Player.
If you are looking for a collaboration on a specific project, I will be happy to help you.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics
- Bioinformatic Tools