dbSNP database for just NA12878

[donquijotes]

So I may be completely wrong here but please correct me since I'm mostly used to doing RNA splicing analysis.

I have sequenced sheared DNA from about 10 NA12878 cells using Illumina and a library prep that uses a non proof reading polymerase so we expect it to introduce lots of errors even early on. What I would like to do is figure out how many false positives it gives me (Allele frequency of >=15%.
To do that I need the NA12878 ref genome, and it's dbSNP database for heterozygous loci. Right? If I just align to NA12878 then the het loci that are endogenous "SNPs" would look as false positives. Where can I find the NA12878 specific dbSNP?

If there is another more logical way of doing this analysis please feel free to call me out.

Thank you!

[donquijotes]

I've actually found this thread that might help.

http://seqanswers.com/forums/showthread.php?t=23093
Two files that I found useful:

ftp://ftp.1000genomes.ebi.ac.uk/vol1...populations.md
shows the population that the vcf files came from. The NA12878 should be the CEU CEPH Utah residents (CEPH) with Northern and Western European ancestry

ftp://ftp.1000genomes.ebi.ac.uk/vol1.../2010_07/trio/
has all the different datasets

If someone else has better options or suggestions please let me know

[sklages]

This might be interesting for you: http://www.illumina.com/platinumgenomes/

[donquijotes]

Thank you sklages! It looks that people update these files frequently so the databases should be way better than the 2010 version the 1000genomes pilot study offers.

[HESmith]

Note that some of your false-positives will be alignment artifacts rather than polymerase errors. If it's important to discriminate those classes, see this paper.

[donquijotes]

HESmith, That was a fantastic paper, thank you for sharing!