If I wanted to use blast:
Do you think 'blastn' would work? With then as -db the human reference genome? My sequences are around 20bp long.
Hi everyone,
I'm trying to find the genomic location of a list of sequences in the reference genome. So if my sequence is 'ACGTACGTAGTCATGC', I want as an output something like this:
chr1 position of the first nucleotide position of the last nucleotide
Are there any tools to do this? Blastn maybe? But since I have the reference genome locally installed, there might be some other options?
3 answers
If you only need exact matches, then give one of the answers in this thread a try: Locating A Sequence In A Fasta File.
Sure, I'd expect so. That's probably faster once the index is made, though you'll need to build the index first. Of course if you only have to do this a couple times, then using one of the methods in the thread I listed is probably faster than building the blast database and filtering the results.
You can also use blat. Since your query sequences are shorter, you might also want to check Using Blat for short sequences with maximum sensitivity
An easy way is to use such mapping tool as bowtie. First you will need to index reference file using bowtie-build.
Than you can find location of your sequence using:
bowtie -c <reference.ebwt.prefix> ACGTACGTAGTCATGC
this will give you coordinates of this sequence.
For more information look at http://bowtie-bio.sourceforge.net/tutorial.shtml
Log in to answer this question.
Do you only want exact matches? You can install blast locally, btw.
Yes indeed, only exact matches. About blast, indeed we have blast locally installed. But I'm not really experienced with blast. So I was wondering if there are other options.