SNP or not

[biocc]

I used MAQ, VAAL and novoalign to map solexa reads on a reference genome . ouput of these softwares showed a delete or a SNP exist. but when I used Tablet for visualization and exploration of sequence assemblies, it seems information from alignment is not sufficient for the mutaion. why? which is right.
thanks.

[NextGenSeq]

IMHO, the only way to validate a SNP is by PCR and first gen sequencing. Obviously you can only do this for a small minority of SNPs.

[Lee Sam]

If you just want to visualize it, you might want to try the UCSC genome browser which allows you to plug in BAM files output from the aligners. I'm not sure how to do it exactly, as someone else set up the infrastructure I'm using.

[ntremblay]

IMHO, the only way to validate a SNP is by PCR and first gen sequencing. Obviously you can only do this for a small minority of SNPs.

You could use a lot of techniques to validate a snp! The traditional way of doing it is PCR + Sanger seq, but new techniques like Taq Man assay, Sequenom Assay, Allelic discrimination with a melting curve in QPCR, etc.!

[drio]

What did you use to call the SNPs? Most tools are not very stringent in the default mode and require tweaking the parameters and/or perform downstream filters to improve the specificity.

[biocc]

thanks for your reply

What did you use to call the SNPs? Most tools are not very stringent in the default mode and require tweaking the parameters and/or perform downstream filters to improve the specificity.

I want to find the genome difference between two bacterial strains and analysis the phenotype of them. I use samtools for downstream filter, is there other methods better? thanks

[severin]

Illumina Goldengate

You could also use Illumina Goldengate technology to verify potential SNPs.
Instead of performing a screen on a small number of SNPs on many genomes, you can perform it on many SNPs for 1 or 2 genomes.

http://www.illumina.com/technology/goldengate_genotyping_assay.ilmn

[drio]

I want to find the genome difference between two bacterial strains and analysis the phenotype of them. I use samtools for downstream filter, is there other methods better? thanks

The samtools SNP calling protocol is a very good starting point. I assume you have a very deep coverage on your samples (since they are bacteria), so you probably want to make the
varfilter parameters more stringent (start increasing the coverage necessary to call a snp). Do you have any expected mutation rate on your bacterias? insertions?

Notice the samtools' author has recently released a new version of samtools and pileup has been deprecated. mpileup is recommend now.